Comprehensive Analysis Identifies Variability in PI3K Pathway Alterations in Triple-Negative Breast Cancer Subtypes

PURPOSE

The PI3K pathway is frequently altered in triple-negative breast cancer (TNBC). Limited cell line and human data suggest that TNBC tumors characterized as mesenchymal (M) and luminal androgen receptor (LAR) subtypes have increased incidence of alterations in the PI3K pathway. The impact of PI3K pathway alterations across TNBC subtypes is poorly understood.

METHODS

Pretreatment tumor was evaluated from operable TNBC patients enrolled on a clinical trial of neoadjuvant therapy (NAT; A Robust TNBC Evaluation fraMework to Improve Survival [ClinicalTrials.gov identifier: NCT02276443]). Tumors were characterized into seven TNBC subtypes per Pietenpol criteria (basal-like 1, basal-like 2, immunomodulatory, M, mesenchymal stem-like, LAR, and unstable). Using whole-exome sequencing, RNA sequencing, and immunohistochemistry for PTEN, alterations were identified in 32 genes known to activate the PI3K pathway. Alterations in each subtype were associated with pathologic response to NAT.

RESULTS

In evaluated patients (N = 177), there was a significant difference in the incidence of PI3K pathway alterations across TNBC subtypes (P < .01). The highest incidence of alterations was seen in LAR (81%), BL2 (79%), and M (62%) subtypes. The odds ratio for pathologic complete response (pCR) in the presence of PIK3CA mutation, PTEN mutation, and/or PTEN loss was highest in the LAR subtype and lowest in the M subtype, but these findings did not reach statistical significance. Presence of PIK3CA mutation was associated with pCR in the LAR subtype (P = .02).

CONCLUSION

PI3K pathway alteration can affect response to NAT in TNBC, and targeted agents may improve outcomes, particularly in patients with M and LAR TNBC.

A phase II study of neoadjuvant atezolizumab and nab-paclitaxel in patients with anthracycline-resistant early-stage triple-negative breast cancer

PURPOSE

Neoadjuvant anti-PD-(L)1 therapy improves the pathological complete response (pCR) rate in unselected triple-negative breast cancer (TNBC). Given the potential for long-term morbidity from immune-related adverse events (irAEs), optimizing the risk-benefit ratio for these agents in the curative neoadjuvant setting is important. Suboptimal clinical response to initial neoadjuvant therapy (NAT) is associated with low rates of pCR (2-5%) and may define a patient selection strategy for neoadjuvant immune checkpoint blockade. We conducted a single-arm phase II study of atezolizumab and nab-paclitaxel as the second phase of NAT in patients with doxorubicin and cyclophosphamide (AC)-resistant TNBC (NCT02530489).

METHODS

Patients with stage I-III, AC-resistant TNBC, defined as disease progression or a < 80% reduction in tumor volume after 4 cycles of AC, were eligible. Patients received atezolizumab (1200 mg IV, Q3weeks × 4) and nab-paclitaxel (100 mg/m² IV,Q1 week × 12) as the second phase of NAT before undergoing surgery followed by adjuvant atezolizumab (1200 mg IV, Q3 weeks, × 4). A two-stage Gehan-type design was employed to detect an improvement in pCR/residual cancer burden class I (RCB-I) rate from 5 to 20%.

RESULTS

From 2/15/2016 through 1/29/2021, 37 patients with AC-resistant TNBC were enrolled. The pCR/RCB-I rate was 46%. No new safety signals were observed. Seven patients (19%) discontinued atezolizumab due to irAEs.

CONCLUSION

This study met its primary endpoint, demonstrating a promising signal of activity in this high-risk population (pCR/RCB-I = 46% vs 5% in historical controls), suggesting that a response-adapted approach to the utilization of neoadjuvant immunotherapy should be considered for further evaluation in a randomized clinical trial.

Predictors of success in establishing orthotopic patient-derived xenograft models of triple negative breast cancer

Patient-derived xenograft (PDX) models of breast cancer are an effective discovery platform and tool for preclinical pharmacologic testing and biomarker identification. We established orthotopic PDX models of triple negative breast cancer (TNBC) from the primary breast tumors of patients prior to and following neoadjuvant chemotherapy (NACT) while they were enrolled in the ARTEMIS trial (NCT02276443). Serial biopsies were obtained from patients prior to treatment (pre-NACT), from poorly responsive disease after four cycles of Adriamycin and cyclophosphamide (AC, mid-NACT), and in cases of AC-resistance, after a 3-month course of different experimental therapies and/or additional chemotherapy (post-NACT). Our study cohort includes a total of 269 fine needle aspirates (FNAs) from 217 women, generating a total of 62 PDX models (overall success-rate = 23%). Success of PDX engraftment was generally higher from those cancers that proved to be treatment-resistant, whether poorly responsive to AC as determined by ultrasound measurements mid-NACT (p = 0.063), RCB II/III status after NACT (p = 0.046), or metastatic relapse within 2 years of surgery (p = 0.008). TNBC molecular subtype determined from gene expression microarrays of pre-NACT tumors revealed no significant association with PDX engraftment rate (p = 0.877). Finally, we developed a statistical model predictive of PDX engraftment using percent Ki67 positive cells in the patient's diagnostic biopsy, positive lymph node status at diagnosis, and low volumetric reduction of the patient's tumor following AC treatment. This novel bank of 62 PDX models of TNBC provides a valuable resource for biomarker discovery and preclinical therapeutic trials aimed at improving neoadjuvant response rates for patients with TNBC.

Quantitative Apparent Diffusion Coefficients From Peritumoral Regions as Early Predictors of Response to Neoadjuvant Systemic Therapy in Triple-Negative Breast Cancer

BACKGROUND

Pathologic complete response (pCR) to neoadjuvant systemic therapy (NAST) in triple-negative breast cancer (TNBC) is a strong predictor of patient survival. Edema in the peritumoral region (PTR) has been reported to be a negative prognostic factor in TNBC.

PURPOSE

To determine whether quantitative apparent diffusion coefficient (ADC) features from PTRs on reduced field-of-view (rFOV) diffusion-weighted imaging (DWI) predict the response to NAST in TNBC.

STUDY TYPE

Prospective.

POPULATION/SUBJECTS

A total of 108 patients with biopsy-proven TNBC who underwent NAST and definitive surgery during 2015-2020.

FIELD STRENGTH/SEQUENCE

A 3.0 T/rFOV single-shot diffusion-weighted echo-planar imaging sequence (DWI).

ASSESSMENT

Three scans were acquired longitudinally (pretreatment, after two cycles of NAST, and after four cycles of NAST). For each scan, 11 ADC histogram features (minimum, maximum, mean, median, standard deviation, kurtosis, skewness and 10th, 25th, 75th, and 90th percentiles) were extracted from tumors and from PTRs of 5 mm, 10 mm, 15 mm, and 20 mm in thickness with inclusion and exclusion of fat-dominant pixels.

STATISTICAL TESTS

ADC features were tested for prediction of pCR, both individually using Mann-Whitney U test and area under the receiver operating characteristic curve (AUC), and in combination in multivariable models with k-fold cross-validation. A P value < 0.05 was considered statistically significant.

RESULTS

Fifty-one patients (47%) had pCR. Maximum ADC from PTR, measured after two and four cycles of NAST, was significantly higher in pCR patients (2.8 ± 0.69 vs 3.5 ± 0.94 mm2 /sec). The top-performing feature for prediction of pCR was the maximum ADC from the 5-mm fat-inclusive PTR after cycle 4 of NAST (AUC: 0.74; 95% confidence interval: 0.64, 0.84). Multivariable models of ADC features performed similarly for fat-inclusive and fat-exclusive PTRs, with AUCs ranging from 0.68 to 0.72 for the cycle 2 and cycle 4 scans.

DATA CONCLUSION

Quantitative ADC features from PTRs may serve as early predictors of the response to NAST in TNBC.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 4.

Molecular Characterization and Prospective Evaluation of Pathologic Response and Outcomes with Neoadjuvant Therapy in Metaplastic Triple-Negative Breast Cancer

PURPOSE

Metaplastic breast cancer (MpBC) is a rare subtype of breast cancer that is commonly triple-negative and poorly responsive to neoadjuvant therapy in retrospective studies.

EXPERIMENTAL DESIGN

To better define clinical outcomes and correlates of response, we analyzed the rate of pathologic complete response (pCR) to neoadjuvant therapy, survival outcomes, and genomic and transcriptomic profiles of the pretreatment tumors in a prospective clinical trial (NCT02276443). A total of 211 patients with triple-negative breast cancer (TNBC), including 39 with MpBC, received doxorubicin-cyclophosphamide-based neoadjuvant therapy.

RESULTS

Although not meeting the threshold for statistical significance, patients with MpBCs were less likely to experience a pCR (23% vs. 40%; P = 0.07), had shorter event-free survival (29.4 vs. 32.2 months, P = 0.15), metastasis-free survival (30.3 vs. 32.4 months, P = 0.22); and overall survival (32.6 vs. 34.3 months, P = 0.21). This heterogeneity is mirrored in the molecular profiling. Mutations in PI3KCA (23% vs. 9%, P = 0.07) and its pathway (41% vs. 18%, P = 0.02) were frequently observed and enriched in MpBCs. The gene expression profiles of each histologically defined subtype were distinguishable and characterized by distinctive gene signatures. Among nonmetaplastic (non-Mp) TNBCs, 10% possessed a metaplastic-like gene expression signature and had pCR rates and survival outcomes similar to MpBC.

CONCLUSIONS

Further investigations will determine if metaplastic-like tumors should be treated more similarly to MpBC in the clinic. The 23% pCR rate in this study suggests that patients with MpBC should be considered for NAT. To improve this rate, a pathway analysis predicted enrichment of histone deacetylase (HDAC) and RTK/MAPK pathways in MpBC, which may serve as new targetable vulnerabilities.

Abstract 1595: Analysis of spatiotemporal phenotypic heterogeneity in chemoresistant triple negative breast cancer using imaging mass cytometry

Shifts in tumor cell phenotype in response to selective pressures (i.e. changing microenvironments, drug treatments) pose one of the biggest obstacles to successful breast cancer therapies. Phenotypically diverse breast tumor and stroma subpopulations, and interactions between them that alter tumor cell biology, represent unique and spatially distinct niches. We hypothesize that localized neighborhoods of breast tumor cells possess specialized phenotypes that mediate chemoresistance and represent novel therapeutic vulnerabilities. In order to assess these potential phenotypes, we utilized imaging mass cytometry (IMC), a highly multiplexed imaging modality that allows simultaneous measurement of 30-40 antigens while retaining the spatial architecture of the cancer tissue. We constructed an IMC antibody panel that combines markers for tissue architecture, tumor and stromal cell phenotyping, and signaling pathway activation. IMC was applied to patient-derived xenograft (PDX) models of triple negative breast cancer (TNBC).Our TNBC PDX collection was established from tumors obtained before and after neoadjuvant Adriamycin and cyclophosphamide (AC). IMC analysis of 18 PDX models representing eight patients revealed that stromal cell phenotypes were generally shared between all models, but tumor cell phenotypes were largely patient-specific. While every model was comprised primarily of a few major tumor cell phenotypes, we noted that each case also harbored several minor, unique populations, suggesting that specialized neighborhoods may exist within the tumor mass. Comparison of paired PDX models showed a wide range of phenotypic responses to chemotherapy, ranging from stable tumor composition to widespread changes in tumor phenotypes. These phenotypic changes arose despite relatively consistent genomic architecture. Vimentinhi fibroblasts were present more often in post-AC models, while SMAhi fibroblasts were unchanged after treatment. Comparison of pre-/post-AC PDX pairs revealed spatially constrained MAPK activation emerged after treatment. To capture acute changes in tumor phenotype, we treated treatment-naïve PDX models with AC and evaluated tumors by IMC. As tumors regressed and then regrew, we identified novel phenotypic shifts, again including increased MAPK signaling localized to discrete neighborhoods, suggesting this property may be a common feature of chemoresistant TNBC. Analysis of adjacent cells revealed seven distinct neighborhoods, and ongoing work is aimed at determining whether these neighborhoods are altered in response to chemotherapy treatment. Taken together, our findings suggest that distinct tumor phenotypes arise following treatment. Our goal is to determine whether these unique phenotypic niches functionally contribute to chemoresistance and if disruption of these niches enhances chemosensitivity.

Citation Format: Amanda L. Rinkenbaugh, Vidya C. Sinha, Pankaj Singh, Yuan Qi, Jiansu Shao, Xiaomei Zhang, Gloria V. Echeverria, W. Fraser Symmans, Stacy L. Moulder, Helen Piwnica-Worms. Analysis of spatiotemporal phenotypic heterogeneity in chemoresistant triple negative breast cancer using imaging mass cytometry [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1595.

Mid-treatment Ultrasound Descriptors as Qualitative Imaging Biomarkers of Pathologic Complete Response in Patients with Triple-Negative Breast Cancer

This study aimed to investigate mid-treatment breast tumor ultrasound characteristics that may predict eventual pathologic complete response (pCR) in triple-negative breast cancer; specifically, we examined associations between pCR and two parameters: tumor response pattern and tumor appearance. Ultrasound was performed at mid-treatment, defined as the completion of four cycles of anthracycline-based chemotherapy and before receiving taxane-based chemotherapy. Consensus imaging review was performed while blinded to pathology results (i.e., pCR/non-pCR) from surgery. Tumor response pattern was described as "complete," "concentric," "fragmented," "stable" or "progression." Tumor appearance was designated as "mass," "architectural distortion," "flat tumor bed" or "clip only." Univariate and multivariate regression analyses of 144 participants showed significant associations between mid-treatment response pattern and pCR (p = 0.0348 and p = 0.0173, respectively), with complete and concentric response patterns more likely to achieve pCR than other patterns. Univariate and multivariate regression analyses further showed significant associations between mid-treatment tumor appearance and pCR (p < 0.0001 for both), with persistent appearance of mass less likely than other appearances to achieve pCR. To conclude, our study demonstrated strong associations between pCR and both tumor response pattern and tumor appearance, thereby suggesting that these parameters have potential as qualitative imaging biomarkers of pCR in triple-negative breast cancer.

Identification of biomarkers of response to preoperative talazoparib monotherapy in treatment naïve gBRCA+ breast cancers

Germline mutations in BRCA1 or BRCA2 exist in ~2–7% of breast cancer patients, which has led to the approval of PARP inhibitors in the advanced setting. We have previously reported a phase II neoadjuvant trial of single agent talazoparib for patients with germline BRCA pathogenic variants with a pathologic complete response (pCR) rate of 53%. As nearly half of the patients treated did not have pCR, better strategies are needed to overcome treatment resistance. To this end, we conducted multi-omic analysis of 13 treatment naïve breast cancer tumors from patients that went on to receive single-agent neoadjuvant talazoparib. We looked for biomarkers that were predictive of response (assessed by residual cancer burden) after 6 months of therapy. We found that all resistant tumors exhibited either the loss of SHLD2, expression of a hypoxia signature, or expression of a stem cell signature. These results indicate that the deep analysis of pre-treatment tumors can identify biomarkers that are predictive of response to talazoparib and potentially other PARP inhibitors, and provides a framework that will allow for better selection of patients for treatment, as well as a roadmap for the development of novel combination therapies to prevent emergence of resistance.

Abstract PD11-07: Integrated model for early prediction of neoadjuvant systemic therapy response in triple negative breast cancer

Background: TNBC constitutes an aggressive and heterogeneous group of tumors with variable response to neoadjuvant therapy (NAT) that currently lacks clinically available profiling strategies for prediction. We aimed to develop an integrated model based on imaging, pathological and clinical data capable to predict NAT response in TNBC early during therapy. METHOD AND MATERIALS:125 Stage I-III TNBC patients enrolled in an IRB approved prospective clinical trial (NCT02276433) who had DCE-MRI at baseline (BL) and post 2 cycles (C2) of NAT, and had surgery were included in this analysis. Tumor volume was calculated using 3D measurements at BL and C2 time points DCE-MRI. Percent tumor volume reduction (TVR) between BL and C2 was calculated. Demographic, clinical, and pathological data (age, T and N stage, histology, androgen receptor expression, Ki-67, stromal tumor infiltrating lymphocytes level (sTIL), and PD-L1 expression), and treatment response at surgery (pCR vs non-pCR) were documented. Recursive partitioning was used to identify TVR cutoff value. Multivariate logistic regression and ROC analysis were used to assess associations and build and evaluate predictive models. RESULTS: 61 (49%) TNBC pts showed pCR at surgery, and 64 (51%) non-pCR. Recursive partitioning analysis identified ≥ 55% TVR as the optimal cutoff values for pCR prediction at C2. TVR, N stage and sTIL were significantly associated with pCR in the multivariate analyses (p&lt;0.002, p&lt;0.01, p&lt;0.001, respectively). Integrated model containing TVR (≥55% vs &lt;55%), N stage (N0 vs N+) and sTIL (≥20% vs &lt;20%) was predictive of pCR with AUC 0.84 (95% CI:0.77-0.91). Integrated model performance was significantly better than TVR only or clinical only (sTIL, and N stage) models (p&lt;0.001). CONCLUSION: Integrated model that included imaging (DCE-MRI TVR), clinical (N stage) and pathological (sTIL) data showed high accuracy for prediction of NAT response in TNBC patients early during treatment. Validation of these results in a large prospective study is ongoing.

Citation Format: Gaiane Margishvili Rauch, Rosalind P. Candelaria, Mary Saber Guirguis, Medine Boge, Rania M. M. Mohamed, Nabil Elshafeey, Jia Sun, Gary J Whitman, Jessica Leung, Huong C Le-Petross, Lumarie Santiago, Deanna Lane, Marion Scoggins, David Spak, Miral M Patel, Frances Perez, Jason B. White, Elizabeth Ravenberg, Wei Peng, Debu Tripathy, Vicente Valero, Jennifer Litton, Lei Huo, Clinton Yam, Alastair Thompson, Jingfei Ma, Stacy L. Moulder, Wei Yang, Beatriz E. Adrada. Integrated model for early prediction of neoadjuvant systemic therapy response in triple negative breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD11-07.

BI-RADS Ultrasound Lexicon Descriptors and Stromal Tumor-Infiltrating Lymphocytes in Triple-Negative Breast Cancer

PURPOSE

Increased levels of stromal tumor-infiltrating lymphocytes (sTILs) have recently been considered a favorable independent prognostic and predictive biomarker in triple-negative breast cancer (TNBC). The purpose of this study was to determine the relationship between BI-RADS (Breast Imaging Reporting and Data System) ultrasound lexicon descriptors and sTILs in TNBC.

MATERIALS AND METHODS

Patients with stage I-III TNBC were evaluated within a single-institution neoadjuvant clinical trial. Two fellowship-trained breast radiologists used the BI-RADS ultrasound lexicon to assess pretreatment tumor shape, margin, echo pattern, orientation, posterior features, and vascularity. sTILs were defined as low <20 or high ≥20 on the pretreatment biopsy. Fisher's exact tests were used to assess the association between lexicon descriptors and sTIL levels.

RESULTS

The 284 patients (mean age 52 years, range 24-79 years) were comprised of 68% (193/284) with low-sTIL tumors and 32% (91/284) with high-sTIL tumors. TNBC tumors with high sTILs were more likely to have the following features: (1) oval/round shape than irregular shape (p = 0.003), (2) circumscribed or microlobulated margins than spiculated, indistinct, or angular margins (p = 0.0005); (3) complex cystic and solid pattern than heterogeneous pattern (p = 0.006); and (4) posterior enhancement than shadowing (p = 0.002). There was no significant association between sTILs and descriptors for orientation and vascularity (p = 0.06 and p = 0.49, respectively).

CONCLUSION

BI-RADS ultrasound descriptors of the pretreatment appearance of a TNBC tumor can be useful in discriminating between tumors with low and high sTIL levels. Therefore, there is a potential use of ultrasound tumor characteristics to complement sTILs when used as stratification factors in treatment algorithms for TNBC.

Neoadjuvant T-DM1/pertuzumab and paclitaxel/trastuzumab/pertuzumab for HER2+ breast cancer in the adaptively randomized I-SPY2 trial

HER2-targeted therapy dramatically improves outcomes in early breast cancer. Here we report the results of two HER2-targeted combinations in the neoadjuvant I-SPY2 phase 2 adaptive platform trial for early breast cancer at high risk of recurrence: ado-trastuzumab emtansine plus pertuzumab (T-DM1/P) and paclitaxel, trastuzumab and pertuzumab (THP). Eligible women have >2.5 cm clinical stage II/III HER2+ breast cancer, adaptively randomized to T-DM1/P, THP, or a common control arm of paclitaxel/trastuzumab (TH), followed by doxorubicin/cyclophosphamide, then surgery. Both T-DM1/P and THP arms 'graduate' in all subtypes: predicted pCR rates are 63%, 72% and 33% for T-DM1/P (n = 52), THP (n = 45) and TH (n = 31) respectively. Toxicity burden is similar between arms. Degree of HER2 pathway signaling and phosphorylation in pretreatment biopsy specimens are associated with response to both T-DM1/P and THP and can further identify highly responsive HER2+ tumors to HER2-directed therapy. This may help identify patients who can safely de-escalate cytotoxic chemotherapy without compromising excellent outcome.

Ganitumab and metformin plus standard neoadjuvant therapy in stage 2/3 breast cancer

I-SPY2 is an adaptively randomized phase 2 clinical trial evaluating novel agents in combination with standard-of-care paclitaxel followed by doxorubicin and cyclophosphamide in the neoadjuvant treatment of breast cancer. Ganitumab is a monoclonal antibody designed to bind and inhibit function of the type I insulin-like growth factor receptor (IGF-1R). Ganitumab was tested in combination with metformin and paclitaxel (PGM) followed by AC compared to standard-of-care alone. While pathologic complete response (pCR) rates were numerically higher in the PGM treatment arm for hormone receptor-negative, HER2-negative breast cancer (32% versus 21%), this small increase did not meet I-SPY's prespecified threshold for graduation. PGM was associated with increased hyperglycemia and elevated hemoglobin A1c (HbA1c), despite the use of metformin in combination with ganitumab. We evaluated several putative predictive biomarkers of ganitumab response (e.g., IGF-1 ligand score, IGF-1R signature, IGFBP5 expression, baseline HbA1c). None were specific predictors of response to PGM, although several signatures were associated with pCR in both arms. Any further development of anti-IGF-1R therapy will require better control of anti-IGF-1R drug-induced hyperglycemia and the development of more predictive biomarkers.

Mammary-specific expression of Trim24 establishes a mouse model of human metaplastic breast cancer

Conditional overexpression of histone reader Tripartite motif containing protein 24 (TRIM24) in mouse mammary epithelia (Trim24COE) drives spontaneous development of mammary carcinosarcoma tumors, lacking ER, PR and HER2. Human carcinosarcomas or metaplastic breast cancers (MpBC) are a rare, chemorefractory subclass of triple-negative breast cancers (TNBC). Comparison of Trim24COE metaplastic carcinosarcoma morphology, TRIM24 protein levels and a derived Trim24COE gene signature reveals strong correlation with human MpBC tumors and MpBC patient-derived xenograft (PDX) models. Global and single-cell tumor profiling reveal Met as a direct oncogenic target of TRIM24, leading to aberrant PI3K/mTOR activation. Here, we find that pharmacological inhibition of these pathways in primary Trim24COE tumor cells and TRIM24-PROTAC treatment of MpBC TNBC PDX tumorspheres decreased cellular viability, suggesting potential in therapeutically targeting TRIM24 and its regulated pathways in TRIM24-expressing TNBC.

PIK3CA mutations in plasma circulating tumor DNA predict survival and treatment outcomes in patients with advanced cancers

Background

Oncogenic mutations in PIK3CA are prevalent in diverse cancers and can be targeted with inhibitors of the phosphoinositide-3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/mTOR) pathway. Analysis of circulating tumor DNA (ctDNA) provides a minimally invasive approach to detect clinically actionable PIK3CA mutations.

Patients and methods

We analyzed PIK3CA hotspot mutation frequency by droplet digital PCR (QX 200; BioRad) using 16 ng of unamplified plasma-derived cell-free DNA from 68 patients with advanced solid tumors (breast cancer, n = 41; colorectal cancer, n = 13; other tumor types, n = 14). Results quantified as variant allele frequencies (VAFs) were compared with previous testing of archival tumor tissue and with patient outcomes.

Results

Of 68 patients, 58 (85%) had PIK3CA mutations in tumor tissue and 43 (74%) PIK3CA mutations in ctDNA with an overall concordance of 72% (49/68, κ = 0.38). In a subset analysis, which excluded samples from 26 patients known not to have disease progression at the time of sample collection, we found an overall concordance of 91% (38/42; κ = 0.74). PIK3CA-mutated ctDNA VAF of ≤8.5% (5% trimmed mean) showed a longer median survival compared with patients with a higher VAF (15.9 versus 9.4 months; 95% confidence interval 6.7-17.1 months; P = 0.014). Longitudinal analysis of ctDNA in 18 patients with serial plasma collections (range 2-22 time points, median 5) showed that those with a decrease in PIK3CA VAF had a longer time to treatment failure (TTF) compared with patients with an increase or no change (10.7 versus 2.6 months; P = 0.048).

Conclusions

Detection of PIK3CA mutations in ctDNA is concordant with testing of archival tumor tissue. Low quantity of PIK3CA-mutant ctDNA is associated with longer survival and a decrease in PIK3CA-mutant ctDNA on therapy is associated with longer TTF.

•

Testing for PIK3CA mutations in ctDNA is concordant with testing of tumor tissue.

•

High PIK3CA-mutant abundance in ctDNA was associated with shorter survival.

•

Increasing PIK3CA-mutant abundance in serial blood samples was associated with shorter TTF.

•

Longitudinal monitoring of PIK3CA-mutant ctDNA tracked with cancer clinical course.

Immune Phenotype and Response to Neoadjuvant Therapy in Triple-Negative Breast Cancer

PURPOSE

Increasing tumor-infiltrating lymphocytes (TIL) is associated with higher rates of pathologic complete response (pCR) to neoadjuvant therapy (NAT) in patients with triple-negative breast cancer (TNBC). However, the presence of TILs does not consistently predict pCR, therefore, the current study was undertaken to more fully characterize the immune cell response and its association with pCR.

EXPERIMENTAL DESIGN

We obtained pretreatment core-needle biopsies from 105 patients with stage I-III TNBC enrolled in ARTEMIS (NCT02276443) who received NAT from Oct 22, 2015 through July 24, 2018. The tumor-immune microenvironment was comprehensively profiled by performing T-cell receptor (TCR) sequencing, programmed death-ligand 1 (PD-L1) IHC, multiplex immunofluorescence, and RNA sequencing on pretreatment tumor samples. The primary endpoint was pathologic response to NAT.

RESULTS

The pCR rate was 40% (42/105). Higher TCR clonality (median = 0.2 vs. 0.1, P = 0.03), PD-L1 positivity (OR: 2.91, P = 0.020), higher CD3⁺:CD68⁺ ratio (median = 14.70 vs. 8.20, P = 0.0128), and closer spatial proximity of T cells to tumor cells (median = 19.26 vs. 21.94 μm, P = 0.0169) were associated with pCR. In a multivariable model, closer spatial proximity of T cells to tumor cells and PD-L1 expression enhanced prediction of pCR when considered in conjunction with clinical stage.

CONCLUSIONS

In patients receiving NAT for TNBC, deep immune profiling through detailed phenotypic characterization and spatial analysis can improve prediction of pCR in patients receiving NAT for TNBC when considered with traditional clinical parameters.

The impact of gut microbial composition on response to neoadjuvant chemotherapy (NACT) in early-stage triple negative breast cancer (TNBC)

590

Background: The impact of gut microbiome on tumor biology, progression and response to immunotherapy has been shown across cancer types. However, there is little known about the impact of gut microbial composition on response to chemotherapy. We have previously shown that the gut microbiome remains unaltered during NACT in a cohort of 32 patients. Here we investigate the association between gut microbiome and response to NACT in a larger cohort of early-stage TNBC. Methods: Longitudinal fecal samples were collected from 85 patients with newly-diagnosed, early-stage TNBC patients enrolled in the ARTEMIS trial (NCT02276443). Patients all received standard NACT with adriamycin/cyclophosphamide (AC); volumetric change was assessed using ultrasound and patients with < 70% volumetric reduction (VR) after 4 cycles of AC were recommended to receive targeted therapy in addition to standard NACT to improve response rates. We performed 16S sequencing on bacterial genomic DNA extracted from 85 pre-AC fecal samples using the 2x250 bp paired-end read protocol. Quality-filtered sequences were clustered into Operational Taxonomic Units and classified using Mothur method with the Silva database version 138. For differential taxa-based univariate analysis, abundant microbiome taxa at species, genus, family, class, and order levels were analyzed using DESeq2 after logit transformation. Alpha-diversity indices within group categories were calculated using phyloseq. Microbial alpha diversity (within-sample diversity) was measured by Simpson's reciprocal index. β-diversity was measured using weighted UniFrac distances between the groups. The association between microbiota abundance and pathologic complete response (pCR) or residual disease (RD) was assessed using DESeq2 analysis. Results: Pre-AC fecal samples from 85 patients were available for analysis. Amongst them, there were 46 patients with pCR and 39 patients with RD. There was no significant difference in alpha diversity (p = 0.8) or beta-diversity (p = 0.7) between the pCR and RD groups. However, relative to patients with RD, the gut microbiome in patients with pCR was enriched for the Bifidobacterium longum species (p = 0.03). The gut microbiome in patients with RD was enriched for Lachnospiraceae (p = 0.03) at the genus level and the Bacteroides thetaiotaomicron species (p = 0.02). Conclusions: We have demonstrated significant differences in the gut microbial composition in patients with pCR as compared to patients with RD. Further investigation in larger studies is needed to support therapeutic exploration of gut microbiome modulation in TNBC patients receiving chemotherapy such as probiotic supplementation or fecal microbiota transplant.

Neoadjuvant atezolizumab (atezo) and nab-paclitaxel (nab-p) in patients (pts) with triple-negative breast cancer (TNBC) with suboptimal clinical response to doxorubicin and cyclophosphamide (AC)

592

Background: Neoadjuvant anti-PD-(L)1 therapy confers an improvement in pathological complete response (pCR) rate in unselected TNBC. However, given the potential for long-term morbidity from immune related adverse events (irAE), it is important to optimize the risk-benefit ratio for the use of these novel agents in the curative neoadjuvant setting. Suboptimal clinical response to neoadjuvant therapy (NAT) by sonography is associated with low rates of pCR rate (2-5%, GeparTrio and Aberdeen trials). Here, we report the results of a single arm phase II study of atezo and nab-p as the second phase of NAT in pts with TNBC with suboptimal clinical response to AC (NCT02530489). Methods: Pts with stage I-III TNBC showing suboptimal response to 4 cycles of doxorubicin and cyclophosphamide (AC), defined as disease progression or a <80% reduction in tumor volume by sonography, were eligible. Pts received atezo (1200mg IV, Q3 weeks x 4), and nab-p (100mg/m2 IV, Q1 week, x 12) as the second phase of NAT before undergoing surgery followed by adjuvant atezo (1200mg IV, Q3 weeks, x 4 cycles). This single arm, two-stage Gehan-type study was designed to detect an improvement in pCR from 5% to 20% in order to deem the regimen worthy of further study in a large, randomized, phase II/III trial; success was defined as pCR in 8 out of 37 pts enrolled. In a subset of pts, sufficient baseline tumor tissue was available for stromal TIL assessment (n=29). Results: 34 pts were enrolled from 2/2016-12/2020. Among the 33 pts who have completed NAT, the pCR rate was 30% (10/33, 95% CI: 16-49%) and the pCR/RCB-I rate was 42% (14/33, 95% CI: 25-61%). Clinicopathological characteristics are described in the table below. Treatment-related adverse events (all grades) occurring in ≥ 20% of pts include fatigue (73%), anemia (55%), peripheral sensory neuropathy (55%), neutropenia (48%), rash (42%), ALT elevation (39%), AST elevation (33%), nausea (30%), anorexia (24%), diarrhea (21%), myalgia (21%). Discontinuation of atezo due to irAEs occurred in 4 pts (12%, nephritis [n=2]; adrenal insufficiency [n=1]; hepatitis [n=1]); 2 of these pts had pCR. Conclusions: This study met its primary endpoint, demonstrating a promising signal of activity in this high risk pt population (pCR=30% vs 5% in historical controls). The 12% discontinuation rate due to irAEs confirms that further evaluation of a strategy administering immunotherapy only to pts with high risk disease not responding to AC warrants further investigation. Exploratory genomic and immunological correlative studies are ongoing. Clinical trial information: NCT02530489. [Table: see text]

High tumor mutation burden fails to predict immune checkpoint blockade response across all cancer types

BACKGROUND

High tumor mutation burden (TMB-H) has been proposed as a predictive biomarker for response to immune checkpoint blockade (ICB), largely due to the potential for tumor mutations to generate immunogenic neoantigens. Despite recent pan-cancer approval of ICB treatment for any TMB-H tumor, as assessed by the targeted FoundationOne CDx assay in nine tumor types, the utility of this biomarker has not been fully demonstrated across all cancers.

PATIENTS AND METHODS

Data from over 10 000 patient tumors included in The Cancer Genome Atlas were used to compare approaches to determine TMB and identify the correlation between predicted neoantigen load and CD8 T cells. Association of TMB with ICB treatment outcomes was analyzed by both objective response rates (ORRs, N = 1551) and overall survival (OS, N = 1936).

RESULTS

In cancer types where CD8 T-cell levels positively correlated with neoantigen load, such as melanoma, lung, and bladder cancers, TMB-H tumors exhibited a 39.8% ORR to ICB [95% confidence interval (CI) 34.9-44.8], which was significantly higher than that observed in low TMB (TMB-L) tumors [odds ratio (OR) = 4.1, 95% CI 2.9-5.8, P < 2 × 10^-16]. In cancer types that showed no relationship between CD8 T-cell levels and neoantigen load, such as breast cancer, prostate cancer, and glioma, TMB-H tumors failed to achieve a 20% ORR (ORR = 15.3%, 95% CI 9.2-23.4, P = 0.95), and exhibited a significantly lower ORR relative to TMB-L tumors (OR = 0.46, 95% CI 0.24-0.88, P = 0.02). Bulk ORRs were not significantly different between the two categories of tumors (P = 0.10) for patient cohorts assessed. Equivalent results were obtained by analyzing OS and by treating TMB as a continuous variable.

CONCLUSIONS

Our analysis failed to support application of TMB-H as a biomarker for treatment with ICB in all solid cancer types. Further tumor type-specific studies are warranted.

Abstract PS3-08: Assessment of early response to neoadjuvant systemic therapy (NAST) of triple-negative breast cancer (TNBC) using chemical exchange saturation transfer (CEST) MRI: A pilot study

Introduction CEST MRI permits quantitation of macromolecules such as amide proteins that are of interest in cancer metabolism. However, optimal CEST acquisition and analysis methods remain undetermined. In this study, we investigated CEST MRI as an imaging biomarker for early treatment response in 51 TNBC patients receiving NAST and compared the performance with two different CEST saturation power levels and two analysis methods.

Methods A total of 51 stage I-III TNBC patients enrolled in the prospective ARTEMIS trial (NCT02276443) had CEST imaging performed on a 3T MRI scanner at baseline before NAST (BL, N = 51), after 2 cycles (C2, N = 37), and 4 cycles (C4, N = 44) of NAST. 33 of the 51 patients had imaging at all 3 time points. 29 of the 33 patients had pathological findings, with N = 16 with pathological complete response (pCR) and N = 13 with non-pCR. Two sets of CEST images using 0.9 and 2.0 µT saturation power levels were acquired and analyzed using the magnetization transfer ratio asymmetry (MTRasym) and the Lorentzian line fitting (Mag3.5) methods, for a total of 4 acquisition/analysis combinations. The group averaged CEST signals, MTRasym at 0.9 and 2.0 µT and Mag3.5 at 0.9 and 2.0 µT, at BL, C2 and C4 were determined and evaluated using unpaired (51 patients) and paired (33 patients) Kruskal-Wallis tests. The Mag3.5 at 0.9 µT and the MTRasym at 2.0 µT were further compared between pCR and non-pCR. The group averaged CEST signals at BL, C2, and C4 were evaluated using the Friedman test for the pCR and the non-PCR groups. Separately, the change in the CEST signal from BL to C2 and C4 was determined for each patient and evaluated using the Mann-Whitney test for both groups. P &lt; 0.05 was considered statistically significant.

Results The MTRasym at BL was higher at 2.0 µT than at 0.9 µT. In contrast, the Mag3.5 at BL was higher at 0.9 µT than at 2.0 µT. The MTRasym at 2.0 µT and the Mag3.5 at 0.9 µT decreased during treatment while the MTRasym at 0.9 µT and the Mag3.5 at 2.0 µT were similar. Both the unpaired and the paired Mag3.5 at 0.9 µT showed a significant decrease at C2 and C4 vs. BL (p &lt; 0.01). The unpaired and paired MTRasym at 2.0 µT showed a decrease, although the change was not significant except for the unpaired data at C4. The decrease in the group averaged Mag3.5 at 0.9 µT was significant at C2 vs. BL for the pCR group (p = 0.04), while it was not significant for the pCR group at C4 vs. BL and for the non-pCR group at either C2 or C4 vs. BL. The group averaged MTRasym at 2.0 µT changes were not significant for either the pCR or the non-pCR groups. None of the CEST signal changes on a per patient basis at C2-BL, C4-BL and C4-C2 were significantly different between the pCR and the non-pCR groups. Further, none of the group averaged CEST signals at BL, C2 and C4 were significantly different between the pCR and the non-pCR groups.

Conclusion Our study demonstrates that the CEST quantitation in TNBC patients undergoing NAST depends on acquisition and analysis. For a maximum change in the CEST effect, Lorentzian line fitting is better paired with acquisition at a low saturation power (0.9 µT) and MTRasym is better paired with acquisition at a high saturation power (2.0 µT). Further, a significant CEST signal decrease was observed in TNBC patients with pCR after NAST when a 0.9 µT saturation power and the Lorentzian line fitting were used. In comparison, the decrease was not significant in non-pCR patients using the same saturation power and analysis method. The results suggest that the CEST signal acquired at 0.9 µT saturation power and analyzed using Lorentzian line fitting may be able to differentiate between pCR and non-pCR among TNBC patients undergoing NAST. Additional studies with a larger patient population are ongoing to further validate our findings and their potential for determining pCR.

Citation Format: Shu Zhang, Gaiane M Rauch, Beatriz E Adrada, Medine Boge, Rania MM Mohamed, Abeer H Abdelhafez, Jong Bum Son, Jia Sun, Nabil A Elshafeey, Jason B White, Deanna L Lane, Jessica WT Leung, Marion E Scoggins, David A Spak, Elsa Arribas, Elizabeth Ravenberg, Lumarie Santiago, Tanya W Moseley, Gary J Whitman, Huong Le-Petross, Benjamin C Musall, Mitsuharu Miyoshi, Xinzeng Wang, Brandy Willis, Stacy Hash, Aikaterini Kotrotsou, Peng Wei, Ken-Pin Hwang, Alastair Thompson, Stacy L Moulder, Rosalind P Candelaria, Wei Yang, Jingfei Ma, Mark D Pagel. Assessment of early response to neoadjuvant systemic therapy (NAST) of triple-negative breast cancer (TNBC) using chemical exchange saturation transfer (CEST) MRI: A pilot study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-08.

Functional Tumor Volume by Fast Dynamic Contrast-Enhanced MRI for Predicting Neoadjuvant Systemic Therapy Response in Triple-Negative Breast Cancer

BACKGROUND

Dynamic contrast-enhanced (DCE) MRI is useful for diagnosis and assessment of treatment response in breast cancer. Fast DCE MRI offers a higher sampling rate of contrast enhancement curves in comparison to conventional DCE MRI, potentially characterizing tumor perfusion kinetics more accurately for measurement of functional tumor volume (FTV) as a predictor of treatment response.

PURPOSE

To investigate FTV by fast DCE MRI as a predictor of neoadjuvant systemic therapy (NAST) response in triple-negative breast cancer (TNBC).

STUDY TYPE

Prospective.

POPULATION/SUBJECTS

Sixty patients with biopsy-confirmed TNBC between December 2016 and September 2020.

FIELD STRENGTH/SEQUENCE

A 3.0 T/3D fast spoiled gradient echo-based DCE MRI ASSESSMENT: Patients underwent MRI at baseline and after four cycles (C4) of NAST, followed by definitive surgery. DCE subtraction images were analyzed in consensus by two breast radiologists with 5 (A.H.A.) and 2 (H.S.M.) years of experience. Tumor volumes (TV) were measured on early and late subtractions. Tumors were segmented on 1 and 2.5-minute early phases subtractions and FTV was determined using optimized signal enhancement thresholds. Interpolated enhancement curves from segmented voxels were used to determine optimal early phase timing.

STATISTICAL TESTS

Tumor volumes were compared between patients who had a pathologic complete response (pCR) and those who did not using the area under the receiver operating curve (AUC) and Mann-Whitney U test.

RESULTS

About 26 of 60 patients (43%) had pCR. FTV at 1 minute after injection at C4 provided the best discrimination between pCR and non-pCR, with AUC (95% confidence interval [CI]) = 0.85 (0.74,0.95) (P < 0.05). The 1-minute timing was optimal for FTV measurements at C4 and for the change between C4 and baseline. TV from the early phase at C4 also yielded a good AUC (95%CI) of 0.82 (0.71,0.93) (P < 0.05).

DATA CONCLUSION

FTV and TV measured at 1 minute after injection can predict response to NAST in TNBC.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY: 4.

Comparison of three FDA-approved diagnostic immunohistochemistry assays of PD-L1 in triple-negative breast carcinoma

The Dako 28-8, Dako 22C3, and Ventana SP142 assays are among the approved programmed death ligand 1 (PD-L1) immunohistochemical companion/complementary diagnostics associated with cancer treatment. To address the concordance of these assays in triple-negative breast cancer (TNBC), we examined PD-L1 expression in 98 TNBC tumors and compared the positive rates using the three assays and three scoring methods: immune cell (IC), tumor cell (TC), and combined tumor cell and immune cell (TCIC) (an equivalent to combined positive score, or CPS). The positive rate for PD-L1 expression with a 1% cutoff was highest with 28-8, followed by the 22C3. These two assays demonstrated almost perfect or substantial agreement in all three scores. There was less agreement between SP142 and the other assays. Using the IC score or the TCIC score at a 1% cutoff (CPS 1), 4% of tumors were positive for PD-L1 with SP142 but negative with the other assays. Using SP142 with a 1% cutoff as a reference, the optimal cutoff for best agreement was at 1% for IC, 30% for TC, and 2% for TCIC (CPS 2) with the other two assays. A 2% cutoff for the 22C3 TCIC (CPS 2) yielded the best agreement with SP142 1% IC cutoff (kappa 0.65). Our study showed the lowest positive rate with SP142 among the three assays. However, the other two assays were not able to identify all tumors that would test positive with SP142 using IC or TCIC/CPS. It is unlikely to achieve high agreement between SP142 and the other two assays by changing the analytical cutoffs.

Abstract PR01: Analysis of spatiotemporal phenotypic heterogeneity in chemoresistant triple negative breast cancer using imaging mass cytometry

Shifts in tumor phenotype in response to selective pressures (i.e. changing microenvironments, drug treatments) pose one of the biggest obstacles to successful cancer therapies. Phenotypically diverse tumor and stroma subpopulations, and interactions between them that alter tumor biology, represent unique signaling niches. We hypothesize that these localized neighborhoods of breast tumor cells possess specialized phenotypes that mediate chemoresistance and represent novel therapeutic vulnerabilities. In order to assess these potential phenotypes, we utilized imaging mass cytometry (IMC), a highly multiplexed imaging modality that allows simultaneous measurement of 30-40 antigens while retaining the spatial architecture of the cancer tissue. We have constructed an IMC antibody panel that combines markers for tissue architecture, tumor and stromal cell phenotyping, and signaling pathway activation. IMC was applied to a collection of patient-derived xenograft (PDX) models of triple negative breast cancer (TNBC). Our PDX collection was established from patient tumors obtained before and after neoadjuvant chemotherapy. IMC analysis of 18 PDX models representing eight patients revealed that stromal cell phenotypes were generally shared between all models, but tumor cell phenotypes were largely patient-specific. While every model was comprised primarily of a few major tumor cell phenotypes, we noted that each case also harbored several minor, unique populations, suggesting that specialized neighborhoods may exist within the tumor mass. Sequential PDX models showed a wide range of phenotypic responses after chemotherapy, ranging from stable tumor composition to widespread changes in tumor phenotypes. Importantly, comparison of multiple pre-/post-chemotherapy PDX pairs revealed spatially constrained MAPK activation emerging after treatment. To capture acute changes in tumor phenotype, we treated animals bearing PDX tumors with chemotherapy and evaluated tumors by IMC. As tumors regressed and then regrew, we identified novel phenotypic shifts, again including increased MAPK signaling localized to discrete neighborhoods, suggesting this phenomenon may be a common feature of chemoresistant TNBC. Taken together, our findings suggest that distinct tumor phenotypes arise following treatment. Ongoing work is examining the spatial arrangement of these cell types to determine their local niche compositions, and then correlating these findings with clinical features. Our goal is to determine whether these unique signaling niches functionally contribute to chemoresistance and if disruption of these niches enhances chemosensitivity.

Citation Format: Amanda L. Rinkenbaugh, Vidya C. Sinha, Jiansu Shao, Xiaomei Zhang, Gloria V. Echeverria, W. Fraser Symmans, Stacy L. Moulder, Helen Piwnica-Worms. Analysis of spatiotemporal phenotypic heterogeneity in chemoresistant triple negative breast cancer using imaging mass cytometry [abstract]. In: Proceedings of the AACR Virtual Special Conference on Tumor Heterogeneity: From Single Cells to Clinical Impact; 2020 Sep 17-18. Philadelphia (PA): AACR; Cancer Res 2020;80(21 Suppl):Abstract nr PR01.

A phase II study of Mirvetuximab Soravtansine in triple-negative breast cancer

Folate receptor alpha (FRα) has been reported to be expressed in up to 80% of triple-negative breast cancers (TNBC) with limited expression in normal tissues, making it a promising therapeutic target. Mirvetuximab soravtansine (mirvetuximab-s) is an antibody drug conjugate which has shown promise in the treatment of FRα-positive solid tumors in early phase clinical trials. Herein, are the results of the first prospective phase II trial evaluating mirvetuximab-s in metastatic TNBC. Patients with advanced, FRα-positive TNBC were enrolled on this study. Mirvetuximab-s was administered at a dose of 6.0 mg/kg every 3 weeks. 96 patients with advanced TNBC consented for screening. FRα staining was performed on tumor tissue obtained from 80 patients. The rate of FRα positivity by immunohistochemistry was 10.0% (8/80). Two patients were treated on study, with best overall responses of stable disease in one and progressive disease in the other. Adverse events were consistent with earlier studies. The study was terminated early due to the low rate of FRα positivity in the screened patient population and lack of disease response in the two patients treated. The observed rate of FRα positivity was considerably lower than previously reported and none of the patients had a partial or complete response. Treatment with mirvetuximab-s should only be further explored in TNBC if an alternate biomarker strategy is developed for patient selection on the basis of additional preclinical data.

Tumor necrosis by pretreatment breast MRI: association with neoadjuvant systemic therapy (NAST) response in triple-negative breast cancer (TNBC)

PURPOSE

To determine if tumor necrosis by pretreatment breast MRI and its quantitative imaging characteristics are associated with response to NAST in TNBC.

METHODS

This retrospective study included 85 TNBC patients (mean age 51.8 ± 13 years) with MRI before NAST and definitive surgery during 2010-2018. Each MRI included T2-weighted, diffusion-weighted (DWI), and dynamic contrast-enhanced (DCE) imaging. For each index carcinoma, total tumor volume including necrosis (TTV), excluding necrosis (TV), and the necrosis-only volume (NV) were segmented on early-phase DCE subtractions and DWI images. NV and %NV were calculated. Percent enhancement on early and late phases of DCE and apparent diffusion coefficient were extracted from TTV, TV, and NV. Association between necrosis with pathological complete response (pCR) was assessed using odds ratio (OR). Multivariable analysis was used to evaluate the prognostic value of necrosis with T stage and nodal status at staging. Mann-Whitney U tests and area under the curve (AUC) were used to assess performance of imaging metrics for discriminating pCR vs non-pCR.

RESULTS

Of 39 patients (46%) with necrosis, 17 had pCR and 22 did not. Necrosis was not associated with pCR (OR, 0.995; 95% confidence interval [CI] 0.4-2.3) and was not an independent prognostic factor when combined with T stage and nodal status at staging (P = 0.46). None of the imaging metrics differed significantly between pCR and non-pCR in patients with necrosis (AUC < 0.6 and P > 0.40).

CONCLUSION

No significant association was found between necrosis by pretreatment MRI or the quantitative imaging characteristics of tumor necrosis and response to NAST in TNBC.

Association of Event-Free and Distant Recurrence-Free Survival With Individual-Level Pathologic Complete Response in Neoadjuvant Treatment of Stages 2 and 3 Breast Cancer: Three-Year Follow-up Analysis for the I-SPY2 Adaptively Randomized Clinical Trial

Importance

Pathologic complete response (pCR) is a known prognostic biomarker for long-term outcomes. The I-SPY2 trial evaluated if the strength of this clinical association persists in the context of a phase 2 neoadjuvant platform trial.

Objective

To evaluate the association of pCR with event-free survival (EFS) and pCR with distant recurrence-free survival (DRFS) in subpopulations of women with high-risk operable breast cancer treated with standard therapy or one of several novel agents.

Design, Setting, and Participants

Multicenter platform trial of women with operable clinical stage 2 or 3 breast cancer with no prior surgery or systemic therapy for breast cancer; primary tumors were 2.5 cm or larger. Women with tumors that were ERBB2 negative/hormone receptor (HR) positive with low 70-gene assay score were excluded. Participants were adaptively randomized to one of several different investigational regimens or control therapy within molecular subtypes from March 2010 through 2016. The analysis included participants with follow-up data available as of February 26, 2019.

Interventions

Standard-of-care neoadjuvant therapy consisting of taxane treatment with or without (as control) one of several investigational agents or combinations followed by doxorubicin and cyclophosphamide.

Main Outcomes and Measures

Pathologic complete response and 3-year EFS and DRFS.

Results

Of the 950 participants (median [range] age, 49 [23-77] years), 330 (34.7%) achieved pCR. Three-year EFS and DRFS for patients who achieved pCR were both 95%. Hazard ratios for pCR vs non-pCR were 0.19 for EFS (95% CI, 0.12-0.31) and 0.21 for DRFS (95% CI, 0.13-0.34) and were similar across molecular subtypes, varying from 0.14 to 0.18 for EFS and 0.10 to 0.20 for DRFS.

Conclusions and Relevance

The 3-year outcomes from the I-SPY2 trial show that, regardless of subtype and/or treatment regimen, including 9 novel therapeutic combinations, achieving pCR after neoadjuvant therapy implies approximately an 80% reduction in recurrence rate. The goal of the I-SPY2 trial is to rapidly identify investigational therapies that may improve pCR when validated in a phase 3 confirmatory trial. Whether pCR is a validated surrogate in the sense that a therapy that improves pCR rate can be assumed to also improve long-term outcome requires further study.

Trial Registration

ClinicalTrials.gov Identifier: NCT01042379.

Abstract 1513: Analysis of spatiotemporal phenotypic heterogeneity in chemoresistant triple negative breast cancer using imaging mass cytometry

Tumors are increasingly appreciated as complex ecosystems, wherein functional interactions between tumor subclones, as well as components of the microenvironment, contribute to progression and drug resistance. While some studies have focused on soluble factors that mediate these interactions, little is known about the communication between physically neighboring tumor subclones (and their microenvironment). To investigate the nature and impact of such localized interactions, we are assessing the activation status of key cancer signaling pathways in neighboring tumor cells; importantly, we are characterizing these features at the single cell level within physically intact tumors. Triple negative breast cancer (TNBC) exhibits a high degree of intratumor heterogeneity, which has contributed to a lack of effective targeted therapy options. Chemotherapy remains the standard of care; however approximately half of patients have substantial residual disease following chemotherapy, which is associated with a high risk of recurrence. Our objective is to spatially define signaling heterogeneity using patient-derived xenograft (PDX) models before and after chemotherapy treatment, to determine if spatially-defined signaling niches drive chemoresistance in TNBC. We hypothesize that neighborhoods of cells possess specialized phenotypes that mediate chemoresistance, and when disrupted, will inhibit the growth of chemoresistant tumors. We are employing imaging mass cytometry (IMC),a next-generation immunostaining approach that allows for simultaneous measurement of 30-40 biomarkers while retaining the spatial organization of the sample.Wehave constructed an IMC panel of antibodies that combines markers for tissue architecture, tumor and immune cell phenotyping, and signaling pathway activation. Our PDX collection features sequential pairs derived from biopsies taken before and after chemotherapy treatment. IMC of the pre-/post-chemotherapy pairs revealed spatial patterns of pathway activation that emerged following treatment, including increases in PI3K/mTOR and localized MAPK signaling. To complement these studies, we treated PDX models with chemotherapy and analyzed tumors via IMC throughout the course of treatment. Again, we observed the emergence of increased MAPK signaling, localized to discrete neighborhoods within the tumor. Taken together, our findings suggest that unique signaling niches arise following treatment. Our goal is to now determine whether these signaling niches functionally contribute to chemoresistance and if disruption of these niches can inhibit the growth of chemoresistant disease.

Citation Format: Amanda L. Rinkenbaugh, Vidya C. Sinha, Gloria V. Echeverria, Xiaomei Zhang, Jiansu Shao, W. Fraser Symmans, Stacy L. Moulder, Helen Piwnica-Worms. Analysis of spatiotemporal phenotypic heterogeneity in chemoresistant triple negative breast cancer using imaging mass cytometry [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1513.

Abstract 1497: Longitudinal response and selection under neoadjuvant systemic therapy (NAST) in triple-negative breast cancer (TNBC): Profiling results from a randomized trial (ARTEMIS; NCT02276443)

Background: The heterogeneity of TNBC results in a spectrum of responses to NAST: 30-40% of patients (pts) have a pathologic complete response (pCR) with an excellent prognosis. Several methods have been used to measure and evaluate residual disease, including ultrasound, MRI scans, histo-pathology. In addition to these, we hypothesize that comprehensive molecular profiling of longitudinal biopsies, with an integrative evaluation of sub-clonal selection and changes in molecular pathways, will serve as a critical biomarker for chemotherapy, and subsequent targeted therapy trials.

Methods: Pts with stage I-III TNBC began a planned 4 cycles of Adriamycin-based chemo (AC). Biopsies were performed pre (mandatory) and post (optional) AC. Volumetric change by ultrasound (VUS) at completion of AC (or progression) was calculated. Pts with sensitive disease received subsequent taxane-based (T) therapy. Pts with insensitive disease were offered phase II trials. Pathologic response was assessed at surgical resection in 85 pts (Training N=55, Validation N=30). Matched samples, pre and post AC (N = 85 pts) underwent transcriptomic and genomic profiling. Samples were classified into six previously identified ARTEMIS subtypes of TNBC (ART-Type) and immune deconvolution and estimation were performed using RNA-Seq profiles. Multiplex IHC using the Vectra platform is being used to validate results from bulk RNASeq experiments. Somatic mutations and copy-number changes were evaluated using, Mutect2, Sequenza (and FACETs), and PhyloWGS (and PyClone).

Results: Predominately, tumors reacted to AC in 4 different patterns with variation in immune and EMT related pathways. Enrichment of EMT (Group 4) was associated with poor prognosis and higher RCB (10.3% vs 42% pCR rates, p&lt;0.05). The global changes in transcription led to ART-Type switching in all subtypes (44% of pts), except LAR subtype. This subtype was enriched in Group 3 (low overall change), and associated with PIK3CA mutations. MYC amplification was more prevalent (40%) in Group 4, associated with higher EMT and poor prognosis than other groups (28%). Multiple time points were leveraged to constrain sub-clonal clustering and enhance the accuracy of phylogenetic tree construction. Significant sub-clonal selection was detected in 22% of evaluable cases with pre and post biopsies (N=55), with analysis of the validation cohort underway. Molecular subtypes were marginally associated with overall and progression-free survival.

Conclusions: Molecular profiling of longitudinal TNBC samples reveals distinct response patterns in tumors and their micro-environments upon treatment with AC. Integrative analysis of genomic and transcriptomic changes can lead to better stratification of response to NAST. These patterns were indicative of pathologic response in the initial cohort (N=55). Analysis of the second cohort (N=30) will be presented as a validation cohort.

Citation Format: Sahil Seth, Lei Huo, Gaiane M. Rauch, Beatriz Adrada, Helen Piwnica-Worms, Bora Lim, Alastair M. Thompson, Elizabeth A. Mittendorf, Timothy Heffernan, Jennifer K. Litton, William F. Symmans, Giulio F. Draetta, Andrew P. Futreal, Jeffrey T. Chang, Stacy L. Moulder. Longitudinal response and selection under neoadjuvant systemic therapy (NAST) in triple-negative breast cancer (TNBC): Profiling results from a randomized trial (ARTEMIS; NCT02276443) [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1497.

Axillary ultrasound during neoadjuvant systemic therapy in triple-negative breast cancer patients

PURPOSE

To investigate the value of performing mid-treatment axillary ultrasound (AUS) in triple-negative breast cancer (TNBC) patients who are undergoing neoadjuvant systemic therapy (NAST) by determining the optimal cutoff number of abnormal nodes associated with residual nodal disease on surgical pathology.

MATERIALS AND METHODS

This sub-study, an interim analysis of an ongoing single-institution clinical trial enrolling patients with stage I-III TNBC, included 106 patients. Number of abnormal nodes at mid-treatment was assessed and recorded by experienced breast radiologists, who empirically categorized lymph nodes using a binary approach of sonographically-normal versus abnormal. Pathologic lymph node positivity was defined as presence of macrometastasis or micrometastasis in ≥1 axillary node from sentinel lymph node biopsy and/or axillary lymph node dissection.

RESULTS

Of 106 patients, 26 (25 %) had residual nodal disease and 80 (75 %) had no nodal disease at surgery. Median number of abnormal nodes at mid-treatment was 5 (standard deviation [SD], 5) for patients with residual nodal disease and 0 (SD, 2) for patients with no nodal disease at surgery (p < 0.0001). TNBC patients with >4 abnormal nodes at mid-treatment had a significantly higher chance of being node-positive at surgery (AUC = 0.908, p < 0.0001; PPV = 90 %).

CONCLUSION

Our data suggest that a cutoff of >4 abnormal nodes on mid-treatment AUS is associated with residual disease post-NAST. If our findings are substantiated by subsequent analyses, then mid-treatment AUS could be used to identify patients unlikely to achieve nodal pathologic complete response and who should be offered alternative therapy.

Comparison of three scoring methods using the FDA-approved 22C3 immunohistochemistry assay to evaluate PD-L1 expression in breast cancer and their association with clinicopathologic factors

BACKGROUND

In the evaluation of PD-L1 expression to select patients for anti-PD-1/PD-L1 treatment, uniform guidelines that account for different immunohistochemistry assays, different cell types and different cutoff values across tumor types are lacking. Data on how different scoring methods compare in breast cancer are scant.

METHODS

Using FDA-approved 22C3 diagnostic immunohistochemistry assay, we retrospectively evaluated PD-L1 expression in 496 primary invasive breast tumors that were not exposed to anti-PD-1/PD-L1 treatment and compared three scoring methods (TC: invasive tumor cells; IC: tumor-infiltrating immune cells; TCIC: a combination of tumor cells and immune cells) in expression frequency and association with clinicopathologic factors.

RESULTS

In the entire cohort, positive PD-L1 expression was observed in 20% of patients by TCIC, 16% by IC, and 10% by TC, with a concordance of 87% between the three methods. In the triple-negative breast cancer patients, positive PD-L1 expression was observed in 35% by TCIC, 31% by IC, and 16% by TC, with a concordance of 76%. Associations between PD-L1 and clinicopathologic factors were investigated according to receptor groups and whether the patients had received neoadjuvant chemotherapy. The three scoring methods showed differences in their associations with clinicopathologic factors in all subgroups studied. Positive PD-L1 expression by IC was significantly associated with worse overall survival in patients with neoadjuvant chemotherapy and showed a trend for worse overall survival and distant metastasis-free survival in triple-negative patients with neoadjuvant chemotherapy. Positive PD-L1 expression by TCIC and TC also showed trends for worse survival in different subgroups.

CONCLUSIONS

Our findings indicate that the three scoring methods with a 1% cutoff are different in their sensitivity for PD-L1 expression and their associations with clinicopathologic factors. Scoring by TCIC is the most sensitive way to identify PD-L1-positive breast cancer by immunohistochemistry. As a prognostic marker, our study suggests that PD-L1 is associated with worse clinical outcome, most often shown by the IC score; however, the other scores may also have clinical implications in some subgroups. Large clinical trials are needed to test the similarities and differences of these scoring methods for their predictive values in anti-PD-1/PD-L1 therapy.

Germline alterations other than BRCA in triple negative breast cancer (TNBC) patients who underwent neoadjuvant therapy (NAT) on a prospective clinical trial

1546

Background: Previous studies have related germline BRCA mutations to pathologic complete response (pCR) in TNBC cohorts. However, prospective data is lacking on the frequency of non- BRCA germline mutations and pCR in TNBC patients who received neoadjuvant therapy (NAT). The aim of this study was to describe germline alterations in comparison with pCR in a prospective cohort of TNBC receiving NAT. Methods: Pre-NAT blood was drawn from patients enrolled in a clinical trial of genomically tailored NAT (ARTEMIS: NCT: 02276443, per eligibility patients had to have negative clinical BRCA tetsing). Germline DNA was extracted and sequenced on a HiSeq4000 sequencer (Illumina, coverage 60X). Reads were aligned to human reference hg19. Variants were filtered against public databases of normal cohorts: esp6500, 1000 genome, ExAC with a frequency cutoff at 1% in any ethnicity. Two integrative scores were used to evaluate the deleteriousness of the missense variants and the variants predicted to be damaging by both scores were included in the analyses. A 105 pan-cancer susceptibility gene panel was selected based on literature data and commercially available gene panels. NAT included anthracycline and taxane based chemotherapy +/- targeted therapy based on tumor genomic expression. Univariate logistic regression models were used to fit pCR for individual mutations, excluding genes mutated in fewer than three patients. All statistical analyses were performed using R version 3.6.1. with a significance of p=0.05. Results: Germline results and pCR were available for 152 patients. Median age was 55 yrs (range: 24-77). 7.9% were stage (st) I, 65.8% st II, 26.3% st III. 55 pts (36%) had pan-cancer associated germline mutations, whereas 33 (21%) had a breast-cancer associated mutation. Greater than 1% mutations were seen in seventeen genes (Table). There was no significant difference in pCR rate after NAT among pts with different germline mutations versus without mutation. Conclusions: Breast cancer related germline mutations other than BRCA in TNBC are relatively common supporting at least a breast panel (not only BRCA1/2) testing. Treatment implications of different germline mutations and their impact on pCR is ongoing on an extended series. [Table: see text]

Prognostic impact of high stromal tumor-infiltrating lymphocytes (sTIL) in the absence of pathologic complete response (pCR) to neoadjuvant therapy (NAT) in early stage triple negative breast cancer (TNBC)

583

Background: Pathologic complete response is an excellent surrogate for disease-free survival (DFS) and overall survival (OS) in TNBC. High sTIL is associated with improved pCR rates in TNBC. Recent data suggest that high sTIL is also associated with improved outcomes in patients who received no chemotherapy for early stage TNBC (Park, Annals of Oncology, 2019). Thus, we hypothesized that high sTIL may have prognostic impact in patients who do not achieve pCR to NAT. Methods: Pretreatment core biopsies from 182 patients with early-stage TNBC enrolled on the ARTEMIS trial (NCT02276443) were evaluated for sTIL by H&E. Patients were stratified according to sTIL (low < 30%, and high > 30%) and pCR (patients with pCR vs. no pCR). The primary outcome measure was DFS, defined from the date of diagnosis to the first local recurrence, distant metastases or death. Cox proportional hazards regression model was used. During follow-up 33 events for DFS were observed. Results: Among subjects who achieve pCR, DFS was excellent regardless of sTIL status and significantly better than those without pCR (p < 0.05). However, patients with high sTIL and no pCR demonstrated significantly worse DFS compared to all subjects having pCR (HR 0.18, 95% CI 0.04-0.76, p = 0.02). Additionally, we did not find a significant difference between high and low sTIL patients who did not achieve pCR. Conclusions: In early TNBC receiving NAT, for patients failing to achieve pCR, high sTIL was not associated with improved DFS; outcomes were comparable to those with low sTIL without pCR. Thus, high sTIL at baseline does not appear to confer an intrinsic prognostic benefit in the absence of pCR.

Statistical modeling of a novel clinical trial design using neoadjuvant therapy (NAT) to personalize therapy in patients (pts) with triple-negative breast cancer (TNBC)

595

Background: 40-50% of pts with TNBC develop pathologic complete response (pCR) with adriamycin/cyclophosphamide (AC)àtaxane (T) NAT; thus, most pts treated in randomized trials (RCTs) adding experimental drugs (ED) to standard NAT do not benefit from trial participation. A personalized trial design that enriches for non-pCR to standard NAT would diminish toxicity from ED in pts who do not need them and enrich ED in high-risk pts that are most likely to benefit. Methods: ARTEMIS (NCT02276443) is a non-randomized trial to study personalization of NAT in TNBC. Tumor biopsies were performed pre-NAT and volumetric change by ultrasound (VCU) after 4 cycles of AC (or upon clinical progression) assessed response. Pts with sensitive TNBC (VCU >=70% after AC) had T as the second phase of NAT. Pts with <70% VCU were offered phase II trials. pCR was assessed at surgical resection. 273 pts had available pCR status and 222 had complete data to generate a model predictive of response using multivariate logistic regression with common clinical factors. Data was randomly divided into training (n=111) and validation (n=111) sets. Results: 85 pts (38%) had pCR and VCU after AC x 4 was the strongest predictor of pCR. Other factors significant on multivariate analysis and included in the model were T stage (T1-4), stromal TIL, Ki67 and PD-L1. When applied to the validation data set, the accuracy of this model for predicting pCR was 76.6%, sensitivity 78.6% and specificity 75.4%. The PPV was 66.0% and the NPV was 85.2% with a ROC curve AUC of 82.4%. Using these data, ED exposure (table) was estimated for the ARTEMIS study design vs a 1:1 or a 2:1 RCT design (with an estimated pCR in control arm=40%), with a demonstrated benefit for personalization. Conclusions: This modeling indicates that personalization of NAT trials has the potential to enrich ED exposure for non-responsive disease as well as diminish ED exposure in pts likely to achieve pCR with standard NAT. Improved prediction of pCR would further enhance personalized trial design. Clinical trial information: NCT02276443 . [Table: see text]

Comprehensive profiling of androgen receptor-positive (AR+) triple-negative breast cancer (TNBC) patients (pts) treated with standard neoadjuvant therapy (NAT) +/- enzalutamide

517

Background: The luminal androgen receptor (LAR) subtype of TNBC has a low pathologic complete response (pCR) rate after NAT. We determined the pCR rate of the enzalutamide and paclitaxel (ZT) regimen for pts with anthracycline-insensitive AR+ TNBC (NCT02689427), and related biomarkers. Methods: ARTEMIS (NCT02276443) is a non-randomized trial to determine if NAT can be used to personalized therapy. Pts received 4 cycles of doxorubicin-based NAT (AC). Pts with insensitive disease by imaging were offered clinical trials as the second phase of NAT based upon molecular profiling of pre-treatment biopsies. Immunohistochemistry (IHC) of AR+≥10% was the threshold for selecting ZT (enzalutamide 160 or 120 mg PO qD + paclitaxel 80 mg/m2 qW for 12 cycles). pCR was determined by surgery after NAT. Trial had two-stage Phase II design, and we report the completed first stage. We evaluated the concordance between Vanderbilt LAR subtype by molecular profiling (microarray and RNAseq) and IHC %AR+ cells. Frequency of PI3K pathway alterations within the LAR subtype was assessed. Results: 267 pts had tumors profiled by IHC, 220 by microarray, 187 by RNAseq and 197 by whole exome sequencing. 96 pts had post-AC RNAseq. LAR scores from both RNAseq and microarray profiling (n = 139) were highly concordant (R = 0.89, P < 0.001) and identified ~10% of TNBCs tested as LAR. The %AR+ cells from IHC correlated with LAR subtype scores according to RNAseq (R = 0.6, P < 0.001), with a cut-point of ≥30% AR+ having the best concordance with LAR subtype. Unlike other subtypes, by serial profiling, LAR TNBCs did not change subtype signatures after exposure to AC. LAR TNBCs had low rates of pCR (23%) and high rates of PI3K pathway activating aberrations (85%); however PI3K aberrations did not correlate with pCR. Seventeen patients with AC-insensitive TNBC received ZT. Five of 15 patients (33.3%) had responses (pCR or RCB-I). Toxicities are Grade (Gr) 4 syncope (n = 1), Gr3 abnormal liver function (n = 2), Gr3 neutropenia (n = 4). IHC & LAR subtype scores did not statistically associate with response to ZT (P = 0.8, P = 0.9). However, all responders to ZT had an upregulated androgen response pathway (ssGSEA Z > 1) as measured by transcriptomic analysis in pre-treatment biopsies analysis (P = 0.05, ppv = 0.56, npv = 1). Conclusions: The LAR TNBC subtype has a low pCR rate to NAT. Among pts with AC-insensitive TNBC, baseline upregulated androgen response pathway and LAR subtype may benefit from the ZT regimen, potentially by PI3K targeting. Clinical trial information: NCT02689427 .

Immune phenotype and response to neoadjuvant systemic therapy (NAST) in triple negative breast cancer (TNBC)

509

Background: In TNBC patients (pts) receiving NAST, increasing tumor infiltrating lymphocytes (TILs) is associated with higher pathologic complete response (pCR) rates. However, since the presence of TIL do not consistently predict pCR, the current study was undertaken to more fully characterize the immune cell response and its association with pCR. Methods: T cell receptor (TCR) sequencing, PD-L1 immunohistochemistry and multiplex immunofluorescence were performed on prospectively collected pre-NAST tumor samples from 98 pts with stage I-III TNBC enrolled in ARTEMIS (NCT: 02276443). TCR clonality was calculated using Shannon’s entropy. PD-L1+ was defined as ≥1% immune cell staining. Response to NAST was defined using the residual cancer burden (RCB) index. Associations between TCR clonality, immune phenotype, and response were examined with the Wilcoxon rank sum test, Spearman’s rank correlation and multivariable logistic regression using stepwise elimination (threshold p > 0.2), as appropriate. Results: The pCR rate was 39% (38/98). pCR was associated with higher TCR clonality (median = 0.2 [in pts with pCR] vs 0.1 [in pts with residual disease], p = 0.05). Notably, the association between pCR and higher TCR clonality was observed in pts with ≥5% TIL (n = 61; p = 0.05) but not in pts with < 5% TIL (n = 37; p = 0.87). Among pts with ≥5% TIL, TCR clonality emerged as the only independent predictor of response in a multivariable model of tumor immune characteristics (odds ratio/0.1 increase in TCR clonality: 3.0, p = 0.021). PD-L1+ status was associated with higher TCR clonality (median = 0.2 [in PD-L1+] vs 0.1 [in PD-L1-], p = 0.004). Higher TCR clonality was associated with higher CD3+ (rho = 0.32, p = 0.0018) and CD3+CD8+ (rho = 0.33, p = 0.0013) infiltration but lower expression of PD-1 on CD3+ (rho = -0.24, p = 0.021) and CD3+CD8+ cells (rho = -0.21, p = 0.037). Conclusions: In TNBC, a more clonal T cell population is associated with an immunologically active microenvironment (higher CD3+ and CD3/8+ T cell; lower PD-1+CD3+ and PD-1+CD3/8+ T cell; PD-L1+) and favorable response to NAST, especially in pts with ≥5% TIL, suggesting a role for deep immune phenotyping in further refining the predictive value of TILs.

Incidence of PI3K pathway alteration and response to neoadjuvant therapy (NAT) in triple negative breast cancer (TNBC) subtypes

561

Background: Limited cell line and human data suggest that TNBCs characterized as mesenchymal and luminal androgen receptor (LAR) commonly have alterations in the PI3K pathway. More data is needed to better characterize the role of the PI3K pathway across TNBC subtypes. Methods: Pre-treatment tumor biopsies were collected from operable TNBC patients (pts) enrolled on a clinical trial of genomically tailored NAT (ARTEMIS; NCT02276443). Tumors were categorized into 5 groups using the Pietenpol criteria: basal-like (BL) comprised of BL-1 and BL-2, mesenchymal and mesenchymal stem-like (M), immunomodulatory (IM), LAR, or unspecified (UNS). Using whole exome sequencing data, variants (single nucleotide polymorphisms and insertions/deletions) and copy number variations (CNVs) were identified in 32 genes known to activate the PI3K pathway. Results: Tumor subtyping and pathologic response to NAT was available in 127 pts (clinical stage I: 9; II: 84; III: 34). PI3K pathway alteration defined as a variant in one of the evaluated genes and/or deletion of PTEN was seen in 76 (60%) tumors. The most frequent alterations were: PTEN deletion (21%), PIK3CA variant (11%), and PIK3R1 variant (8%). PI3K alteration and residual cancer burden (RCB) rates across TNBC subtypes are shown in the table. There was a significant difference in pathologic complete response (pCR)/RCB 0 rate after NAT across TNBC subtypes (chi2 test; P = 0.02). There was a significant difference in the incidence of PI3K pathway alteration across TNBC subtypes (chi2 test; P < 0.01). Overall, the presence of PI3K alteration was not associated with pCR (Fisher exact test; P = 0.85). Pts with M tumors had a higher rate of substantial residual disease (RCB II-III) after NAT. Presence of PI3K pathway alteration was common in the M subtype and associated with RCB II-III (82% in PI3K-altered vs 33% in wild-type tumors; Fisher exact test; P = 0.02). Presence of PI3K pathway alteration was common but not associated with response in the LAR subtype. Conclusions: The incidence of PI3K pathway alteration varied by TNBC subtype but was not associated with pathologic response to NAT with the exception of increased substantial residual disease (RCB II-III) in the M subtype. [Table: see text]

Resistance to neoadjuvant chemotherapy in triple-negative breast cancer mediated by a reversible drug-tolerant state

Eradicating triple-negative breast cancer (TNBC) resistant to neoadjuvant chemotherapy (NACT) is a critical unmet clinical need. In this study, patient-derived xenograft (PDX) models of treatment-naïve TNBC and serial biopsies from TNBC patients undergoing NACT were used to elucidate mechanisms of chemoresistance in the neoadjuvant setting. Barcode-mediated clonal tracking and genomic sequencing of PDX tumors revealed that residual tumors remaining after treatment with standard frontline chemotherapies, doxorubicin (Adriamycin) combined with cyclophosphamide (AC), maintained the subclonal architecture of untreated tumors, yet their transcriptomes, proteomes, and histologic features were distinct from those of untreated tumors. Once treatment was halted, residual tumors gave rise to AC-sensitive tumors with similar transcriptomes, proteomes, and histological features to those of untreated tumors. Together, these results demonstrated that tumors can adopt a reversible drug-tolerant state that does not involve clonal selection as an AC resistance mechanism. Serial biopsies obtained from patients with TNBC undergoing NACT revealed similar histologic changes and maintenance of stable subclonal architecture, demonstrating that AC-treated PDXs capture molecular features characteristic of human TNBC chemoresistance. Last, pharmacologic inhibition of oxidative phosphorylation using an inhibitor currently in phase 1 clinical development delayed residual tumor regrowth. Thus, AC resistance in treatment-naïve TNBC can be mediated by nonselective mechanisms that confer a reversible chemotherapy-tolerant state with targetable vulnerabilities.

Abstract P6-02-03: Quantitative apparent diffusion coefficient (ADC) radiomics of tumor and peritumoral regions as potential predictors of treatment response to neoadjuvant chemotherapy (NACT) in triple negative breast cancer (TNBC) patients

Background and Purpose: TNBC is comprised of biologically aggressive tumors with diverse clinical behavior and response to chemotherapy. Prediction of disease response to NACT is critical to the development of personalized medicine in TNBC. We evaluated first-order radiomic features from quantitative ADC maps of the tumor and peritumoral region as discriminators of response to NACT in TNBC patients.

Materials and Methods: This IRB-approved prospective study (ARTEMIS trial, NCT02276443) included 34 patients with biopsy proven stage I-III TNBC who underwent evaluation of treatment response by multi-parametric MRI. Patients had a baseline MRI (BL) and a second MRI after 4 cycles (C4) of their treatment. After completion of NACT, all patients underwent surgery and were classified as pathologic complete response (pCR) or non-pCR.

Both MRI exams included T2W series, a dynamic contrast enhanced series (DCE), a conventional diffusion weighted imaging (DWI) series, and a reduced field of view (rFOV) DWI series. Tumor volumes were contoured by an experienced breast radiologist on ADC maps with reference to b1000 DWI images. Regions with necrosis or clip artifacts were excluded from the contour. Peritumoral regions were defined as a 5 mm rim of tissue surrounding the tumor based on DCE series, T2-weighted images with fat suppression and ADC maps. Thirteen first-order radiomic features, including mean, minimum, maximum, percentiles, kurtosis and skewness at a single measurement and the difference between BL and C4 were compared between pCR and non-pCR using Receiver Operating Characteristic (ROC) curve and Wilcoxon rank sum test.

Results: The kurtosis of tumor at C4 by conventional DWI was significantly higher in non-pCR than in pCR patients (AUC=0.785, p=0.0097). The change in kurtosis from BL to C4 by conventional DWI was also significantly higher in non-pCR than in pCR patients (AUC=0.73, p=0.043). The skewness of tumor at C4 by rFOV DWI scan was significantly lower in pCR than non-pCR patients (AUC=0.73, p=0.023).

The 10th percentile of the peritumoral region’s ADC was significantly different between pCR and non-pCR (mean=1.19, SD is ± 0.27 10-3 mm2/s vs mean=1.34, SD ± 0.27 10-3 mm2/s respectively, AUC=0.70, p=0.048). The kurtosis and 25th percentile of the ADC of peritumoral region were borderline significantly different between pCR and non-pCR (AUC=0.69, p=0.067; AUC=0.69, p= 0.073 respectively).

Conclusion: ADC first-order radiomic features from tumor and peritumoral region in TNBC may be useful for predicting treatment response to NACT. Larger study is necessary and is currently in progress to validate these findings.

Citation Format: Beatriz E. Adrada, Abeer H. Abdelhafez, Benjamin C. Musall, Kenneth R. Hess, Jong Bum Son, Mark D. Pagel, Ken-Pin Hwang, Rosalind P. Candelaria, Lumarie Santiago, Gary J. Whitman, Huong Le-Petross, Tanya W. Moseley, Elsa Arribas, Deanna L. Lane, Marion E. Scoggins, David A. Spak, Jessica W.T. Leung, Senthil Damodaran, Bora Lim, Vicente Valeo, Jason B White, Alastair M. Thompson, Jennifer K. Litton, Stacy L. Moulder, Jingfei Ma, Wei T. Yang, Gaiane M Rauch. Quantitative apparent diffusion coefficient (ADC) radiomics of tumor and peritumoral regions as potential predictors of treatment response to neoadjuvant chemotherapy (NACT) in triple negative breast cancer (TNBC) patients [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P6-02-03.

Abstract GS4-02: Investigating genomic and phenotypic evolution of triple negative breast cancer chemoresistance and metastasis in patient-derived xenografts

Approximately 50% of patients with newly diagnosed triple negative breast cancer (TNBC) will have substantial residual cancer burden following neoadjuvant chemotherapy (NACT), resulting in distant metastasis and death for most patients. While intra-tumor heterogeneity (ITH) is pervasive in TNBC, the functional contributions of heterogeneous tumor cell populations to resistance and metastasis remain unclear.

To investigate tumor evolution, we employed orthotopic patient-derived xenograft (PDX) models of treatment-naïve TNBC. A subset of PDX models exhibited partial tumor regression following standard front-line NACT, but repopulated tumors with chemo-sensitive cells after a drug holiday, suggesting that resistance may be mediated by a plastic state. Cellular barcoding and genomic sequencing revealed that residual tumors entered a transient chemotherapy-tolerant phenotypic state not mediated by clonal selection. Altered tumor cell metabolism was a functional vulnerability of residual tumor cells. Residual tumors exhibited heightened mitochondrial load and oxidative capacity compared to pre-treatment tumors. Furthermore, treatment with IACS-010759, a small molecule inhibitor of electron transport chain Complex I currently in phase I clinical development, significantly delayed the regrowth of residual tumors. Features of the residual tumor state were also observed in serial biopsies obtained pre- and post-AC from TNBC patients (NCT02276443). While the mechanisms contributing to altered mitochondrial metabolism in chemoresistant TNBCs remain unclear, preliminary findings suggest that altered mitochondrial dynamics may contribute to the enhanced dependence on oxidative phosphorylation in residual tumors. Collectively, these studies reveal that a reversible phenotypic state characterized by altered tumor cell metabolism can confer chemoresistance and that the residual tumor state may be a novel therapeutic window for chemoresistant TNBC.

NACT resistance leads to distant metastasis, often to multiple organs, in most TNBC patients. However, the relatedness of metastases across diverse secondary sites is not well understood. To model the metastatic cascade, sub-lines of orthotopic PDX models harboring a bioluminescent label were generated. Cellular barcoding and genomic sequencing analyses were conducted on primary mammary tumors and lung, liver, and brain metastases from PDX models. Only a minority of primary tumor clones were detected in metastases, indicating that a selective bottleneck had occurred. While each metastatic lesion harbored numerous low-abundance barcoded lineages, only a select few (&lt;10) outgrew and were predominant. Interestingly, the exact same barcoded lineages predominated in lung, liver, and brain metastases. To delineate the transcriptomic profiles of metastatic subclones, single-cell RNA sequencing analyses are being conducted on primary tumors and multi-organ metastases from PDX models. These studies have revealed transcriptomic ITH in primary and metastatic tumors, with stable patterns of transcriptomic ITH in spatially distinct metastases. Furthermore, metastases exhibited reproducible enrichment of a low-abundance primary tumor transcriptomic subpopulation. Together, these studies will elucidate transcriptomic programs associated multi-organ metastasis in TNBC and are expected to enable rational therapeutic targeting strategies.

Citation Format: Gloria V Echeverria, Mingchu Xu, Jiansu Shao, Xiaomei Zhang, Sabrina Jeter-Jones, Xinhui Zhou, Stacy L Moulder, Joseph R Marszalek, Timothy P Heffernan, Fraser W Symmans, Jeffrey T Chang, Helen Piwnica-Worms. Investigating genomic and phenotypic evolution of triple negative breast cancer chemoresistance and metastasis in patient-derived xenografts [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr GS4-02.

Blocking c-Met and EGFR reverses acquired resistance of PARP inhibitors in triple-negative breast cancer

The limited treatment options and therapeutic failure due to acquired resistance for patients with triple-negative breast cancer (TNBC) represent a significant challenge. Inhibitors against poly (ADP-ribose) polymerase (PARP), olaparib and talazoparib, were recently approved for the treatment of metastatic breast cancer (including TNBC) in patients with germline BRCA1/2 mutations. Despite impressive response rates of ~60%, the prolongation in median progression-free survival with a PARPi is modest, suggesting the emergence of resistance. Several studies have reported that receptor tyrosine kinases (RTKs), such as c-MET (also known as hepatocyte growth factor receptor), are involved in resistance to various anti-neoplastic agents, including PARPi. However, the mechanism by which c-MET contributes to acquired resistance to PARPi in TNBC is not fully understood. In this study, we show that hyperactivated c-Met is detected in TNBC cells with acquired resistance to PARPi, and the combination of talazoparib and crizotinib (a multi-kinase inhibitor that inhibits c-MET) synergistically inhibits proliferation in these cells. Unexpectedly, depleting c-MET had limited effect on talazoparib sensitivity in PARPi-resistant cells. Interestingly, we found evidence of epidermal growth factor receptor (EGFR) hyperactivation and interaction of EGFR/c-Met in these cells. Notably, combining EGFR and PARP inhibitors resulted in greater inhibition of proliferation in c-MET-depleted TNBC cells, and combined c-MET and EGFR inhibition increased sensitivity to talazoparib in TNBC cells with acquired resistance to PARPi. Our findings suggest that combined inhibition of c-MET and EGFR could potentially re-sensitize TNBC to the cytotoxic effects of PARPi.

Neoadjuvant Talazoparib for Patients With Operable Breast Cancer With a Germline BRCA Pathogenic Variant

PURPOSE

Talazoparib has demonstrated efficacy in patients with BRCA-positive metastatic breast cancer. This study evaluated the pathologic response of talazoparib alone for 6 months in patients with a known germline BRCA pathogenic variant (gBRCA-positive) and operable breast cancer.

METHODS

Eligibility included 1 cm or larger invasive tumor and gBRCA-positive disease. Human epidermal growth factor receptor 2-positive tumors were excluded. Twenty patients underwent a pretreatment biopsy, 6 months of once per day oral talazoparib (1 mg), followed by definitive surgery. Patients received adjuvant therapy at physician's discretion. The primary end point was residual cancer burden (RCB). With 20 patients, the RCB-0 plus RCB-I response rate can be estimated with a 95% CI with half width less than 20%.

RESULTS

Twenty patients were enrolled from August 2016 to September 2017. Median age was 38 years (range, 23 to 58 years); 16 patients were gBRCA1 positive and 4 patients were gBRCA2 positive. Fifteen patients had triple-negative breast cancer (estrogen receptor/progesterone receptor < 10%), and five had hormone receptor-positive disease. Five patients had clinical stage I disease, 12 had stage II, and three had stage III, including one patient with inflammatory breast carcinoma and one with metaplastic chondrosarcomatous carcinoma. One patient chose to receive chemotherapy before surgery and was not included in RCB analyses. RCB-0 (pathologic complete response) rate was 53% and RCB-0/I was 63%. Eight patients (40%) had grade 3 anemia and required a transfusion, three patients had grade 3 neutropenia, and 1 patient had grade 4 thrombocytopenia. Common grade 1 or 2 toxicities were nausea, fatigue, neutropenia, alopecia, dizziness, and dyspnea. Toxicities were managed by dose reduction and transfusions. Nine patients required dose reduction.

CONCLUSION

Neoadjuvant single-agent oral talazoparib once per day for 6 months without chemotherapy produced substantial RCB-0 rate with manageable toxicity. The substantive pathologic response to single-agent talazoparib supports the larger, ongoing neoadjuvant trial (ClinicalTrials.gov identifier: NCT03499353).

Abstract 2901: Clonal dynamics and phenotypic evolution during chemoresistance and metastasis revealed by patient-derived xenograft models of triple negative breast cancer

Half of all triple negative breast cancer (TNBC) patients harbor significant residual cancer burden following standard neoadjuvant chemotherapy treatment, resulting in distant metastasis and death for most of these patients. Intra-tumor heterogeneity (ITH) is pervasive in TNBC as is a barrier to development of effective therapeutic strategies, but the relative contributions of heterogeneous tumor cell populations to chemoresistance and metastasis are not well understood. To investigate the clonal dynamics that accompany chemotherapy treatment and metastasis, we employed orthotopic patient-derived xenograft (PDX) models of treatment-naïve TNBC, thus enabling experimentation with heterogeneous populations of human tumor cells that have undergone minimal manipulation.

To monitor the fates of PDX tumor cell lineages as they metastasized, we introduced a pooled lentiviral barcode library (Cellecta) into freshly dissociated PDX tumor cells which were orthotopically engrafted into recipient mice. Genomic analyses, including barcode enumeration, whole-exome sequencing, custom targeted DNA sequencing, and transcriptome sequencing, were conducted to characterize the clonal dynamics that accompanied metastasis. Of the thousands of diverse primary tumor clones, only ~2% harbored metastatic capacity. Of those, a rare population of the exact same clones predominated metastases in lung, liver, and brain, the three most common sites of human TNBC metastasis. These studies provide a quantitative map of the clonal architecture of multi-organ metastasis in TNBC and reveal that identical subclones can thrive in diverse secondary organ microenvironments.

NACT resistance leads to metastasis and death for most patients, yet the origins of chemoresistance in TNBC are unclear. We modeled NACT resistance in an array of PDX models derived from treatment-naïve TNBC biopsies in alignment with an ongoing neoadjuvant clinical trial (NCT02276443). Upon partial response to NACT, tumors entered a transient drug-tolerant state characterized by distinct histologic, proteomic, and transcriptomic features that were reverted as tumors regrew after cessation of treatment. Barcode-mediated lineage tracking and whole-exome sequencing revealed that the drug-tolerant state was not mediated by clonal selection. Based on transcriptomic and metabolic features of the drug-tolerant state, we conducted preclinical trials with an inhibitor of mitochondrial oxidative phosphorylation (IACS-010759), which significantly delayed the regrowth of residual tumors in PDX models. Together, these studies revealed that TNBCs can resist NACT through non-selective adaptation of a reversible phenotypic state, and that inhibition of oxidative phosphorylation may be a promising therapy in the neoadjuvant setting for TNBC.

Citation Format: Gloria V. Echeverria, Zhongqi Ge, Sahil Seth, Emily Powell, Xiaomei Zhang, Sabrina Jeter-Jones, Xinhui Zhou, Yan Jiang, Aaron McCoy, Shirong Cai, Yizheng Tu, Michael Peoples, Yuting Sun, Huan Qiu, Christopher Bristow, Alessandro Carugo, Jiansu Shao, Stacy L. Moulder, William F. Symmans, Timothy P. Heffernan, Jeffrey T. Chang, Helen M. Piwnica-Worms. Clonal dynamics and phenotypic evolution during chemoresistance and metastasis revealed by patient-derived xenograft models of triple negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2901.

Evaluation of predictive biomarkers for AR therapy and to identify the LAR subtype of TNBC

595

Background: Androgen-receptor-like (LAR) triple-negative breast cancer (TNBC) is a subtype identified using Vanderbilt’s molecular signature. LAR subtype has the lowest pCR rate for NACT among all TNBC subtypes (10% vs. 28% for TNBC in general). We launched a clinical trial to determine the effectiveness of enzalutamide and paclitaxel (ZT) in improving this poor chemo. response in the neoadjuvant setting for pts with anthracycline-refractory, androgen receptor (AR)+ TNBC (NCT02689427). However, we do not yet have a robust predictive biomarker to detect an activated AR pathway and have not seen a robust correlation between molecular LAR subtype and AR IHC staining intensity. Methods: Molecular profiling and immunohistochemical analysis of key biomarkers (LAR, Ki67, and vimentin) was performed for all pts enrolled in A Randomized triple negative breast cancer enrolling Trial to Confirm Molecular Profiling Improves Survival (ARTEMIS; NCT02276443). Patients receive 4 cycles of AC, followed by an experimental arm or standard taxane, tailored using nuclear IHC staining. IHC staining of ≥30% AR+ was used as a threshold for selection for enzalutamide combination arm. We evaluated the concordance between LAR-subtype using molecular profiling vs % AR+ cells via IHC. Results: As part of the clinical trial, tumors with ≥30% AR+ cells were classified as LAR. In addition, we used RNA profiling to assign Vanderbilt subtype scores, resulting in classification of 15 tumors as LAR+. We observed a significant correlation (r=0.75) between LAR score and %AR+ cells, with 13 of 15 LAR tumors having ≥30% AR+ cells. Among patients with high % of AR+ tumor cells, 11 received enzalutamide, with 43% (3/7) having responses (pCR or RCB-I). Conclusions: Comparison on numerical scores for Vanderbilt subtype and IHC scores suggests ≥30% AR+ IHC staining as the threshold (ppv=0.65, npv=0.98, Table) to identify the molecular LAR subtype. We observed a trend where response rate was higher in patients with ≥ AR+ IHC scores treated with enzalutamide; however, these results need confirmation in a larger cohort of patients. Clinical trial information: NCT02689427, NCT02276443. [Table: see text]

Pathologic complete response and serial circulating tumor cell monitoring during neoadjuvant therapy in nonmetastatic triple-negative breast cancer patients

e12111

Background: ARTEMIS (A Randomized, TNBC Enrolling trial to confirm Molecular profiling Improves Survival) is a randomized trial to determine if precision guided neoadjuvant chemotherapy (NAC) influences rates of pathologic complete response in the breast and axillary nodes (pCR). We hypothesized that blood circulating tumor cell (CTC) identification before and/or after NAC (just before surgery) would be associated with pCR in non-metastatic TNBC patients. Methods: Blood was assessed for CTCs at baseline (pre-NAC) and after NAC (post-NAC), as part of an IRB approved study, ARTEMIS (2014 – 0185/PA15-1050). CTCs (per 7.5 ml blood) were identified using the Cell Search System (Menarini Silicon Biosystems). Samples > one cell having morphologic criteria for malignancy were deemed CTC+. Fisher exact test was used to evaluate associations between CTC detection and clinicopathologic characteristics, and Chi2 test was used to analyze associations between CTC detection pre- and post- NAC, and pCR. Results: Fifty-nine patients had pre-NAC and 43 patients had post-NAC CTC assessments. One or more CTC was detected in 13/59 (22%) patients pre-NAC and in 16/43 (37%) of post NAC samples. CTC detection was not associated with patient stage, tumor androgen receptor expression, vimentin expression, Ki-67 staining, or tumor-infiltrating lymphocyte levels. Pre-NAC CTC detection was not (chi2 P = 0.98), but post-NAC (chi2 P = 0.0017) CTC identification was negatively associated with pCR. Conclusions: CTC detection post-NAC was negatively associated with pCR in non-metastatic TNBC patients. These findings warrant larger studies studying serial CTC detection throughout treatment to assess response to NAC.

Triplet therapy (continuous ribociclib, everolimus, exemestane) in HR+/HER2− advanced breast cancer postprogression on a CDK4/6 inhibitor (TRINITI-1): Efficacy, safety, and biomarker results

1016

Background: The combination of CDK4/6 inhibitor (CDK4/6i) + endocrine therapy (ET) provides consistent improvement in PFS and response rates compared with single-agent ET as first- or subsequent-line therapy in HR+, HER2− advanced breast cancer (ABC), but the optimal regimen postCDK4/6i progression, including the role of continued CDK 4/6 blockade, is unclear. Methods: TRINITI-1 is a Phase I/II, open-label trial (NCT02732119) of triplet therapy: ribociclib (RIB; CDK4/6i) + everolimus (EVE; mTORi) + exemestane (EXE; ET) in men or postmenopausal women with HR+, HER2− ABC that progressed on prior CDK4/6i and up to 3 lines of therapy (≥ 1 ET and ≤ 1 chemotherapy regimen). Phase I determined RP2D; Phase II assessed efficacy/safety of RIB 300 or 200 mg + EVE 2.5 or 5 mg + EXE 25 mg/day. Here we present the first results in the entire patient population who received this triplet regimen and the correlation of biomarkers with outcomes. Results: As of October 24, 2018, 95 patients were evaluable (ET refractory and postCDK4/6i) in Phases I (n = 17) and II (n = 78). Continuous RIB + EVE + EXE demonstrated clinical benefit at week 24 in 39 patients (41.1%), exceeding the predefined primary end point threshold (> 10%). ORR was 8.4% by investigator assessment, median PFS was 5.7 months, and 1-year PFS was 33%. AEs were consistent with known safety profile of RIB, EVE, and EXE. Most common AEs were neutropenia (all grades, 41.7%; grade 3/4, 31.3%), stomatitis (41.7%; 3.1%), and fatigue (35.4%; 1.0%). No grade 3/4 QTc prolongation was noted. ctDNA genotyping revealed patients with certain tumor alterations, eg ESR1, had shorter median PFS vs wild-type: 3.5 vs 6.9 mo (HR 1.76, 95% CI 1.01–3.05). Additional genomic results, including PIK3CA, will be presented. Conclusions: TRINITI-1 met its primary efficacy end point and is the first trial to demonstrate clinical benefit and tolerability of continuous triplet therapy with ET + mTORi + CDK4/6i in patients with ET-refractory HR+, HER2− ABC postCDK4/6i progression. Tumor genomic profile might impact the clinical outcome with triplet therapy and warrants additional research to guide rational therapy selection. Clinical trial information: NCT02732119.

HER family protein expression and activation predicts response to combination T-DM1/pertuzumab in HER2+ patients in the I-SPY 2 TRIAL

3133

Background: T-DM1 (T), a conjugate of the anti-HER2 therapeutic antibody trastuzumab and the microtubule assembly inhibitor emtansine, was administered in combination with pertuzumab (P), an anti-HER2 therapeutic antibody, to HER2+ breast cancer patients in the I-SPY 2 TRIAL, and graduated in all HER2+ subtypes. Pre-specified biomarker analysis was performed to identify candidate biomarkers associated with pCR within the HER family and cell proliferation pathways in patients treated with T+P. We hypothesized that quantitative measurement and activation of HER2 and activation of its major dimerization partner, EGFR would predict response to T+P. Methods: In the T+P treatment arm, 49 had RPPA and pCR data. 40 RPPA endpoints including 14 total/phospho-proteins in the HER family were assessed for association with pCR using logistic regression (likelihood ratio test; p < 0.05). Analysis was also performed adjusting for HR status and within HR subsets. Markers were analyzed individually; multiple comparison correction (Benjamini-Hochberg) was applied to all p-values. Our statistics are descriptive and do not adjust for multiplicities of other biomarkers outside this study. Results: Of the endpoints tested, only quantitative total HER2 expression, phospho-HER2 (Y1248 and Y877), phospho-EGFR (Y1173 and Y1068), and phospho-SHC Y317 had a positive association with response in the population as a whole, and in a model adjusting for HR status (BH p < 0.05). In HR subset analysis, these 5 analytes had uncorrected p < 0.05 regardless of HR subtype but only survived p-value correction in HR+ tumors. Conclusions: Quantitative measurement of HER2 protein positively associates with response to T+P in patients already identified as HER2+ by central IHC and FISH testing. Activation of HER2 and its dimerization partner, EGFR, also associate with response to T+P in HR+ patients. While our results need to be validated in larger prospective trials, they indicate that new approaches to measure more quantitatively the amount and activation state of HER2 and activated EGFR may more effectively identify patients that respond to HER2 targeted therapies than HER2 IHC and FISH alone.

Beyond TILs: Predictors of pathologic complete response (pCR) in triple-negative breast cancer (TNBC) patients with moderate tumor-infiltrating lymphocytes (TIL) receiving neoadjuvant therapy

572

Background: Increased TIL in TNBC is associated with higher rates of pCR. High TIL is also associated with improved disease free survival and overall survival. The aim of this study is to identify data cut-points of pre-treatment low, moderate and high TIL count based on pCR and to identify clinical and pathological predictors of pCR in patients with moderate TIL. Methods: We evaluated the relationship between pCR and TIL in 180 patients with stage I-III TNBC enrolled in the ARTEMIS trial (NCT02276443). Recursive portioning was used to identify cut-points. Clinical and pathological variables such as age at diagnosis, stage, race, histology as well as Ki-67, vimentin, and androgen receptor (AR) by immunohistochemistry, were evaluated in pts with moderate TIL. A multivariable logistic regression model identified variables independently, significantly associated with pCR. Results: Four TIL groups were identified with pCR rates of 23%, 31%, 48% and 78% respectively (p < 0.0001) (Table A). In the two combined moderate TIL groups, 90 (97%) pts were evaluable for the multivariate model. Stage I-II disease, high Ki-67 and low AR were associated with increased probability of pCR (Table B). The multivariable logistic regression model area under the ROC curve was 0.78 (95% CI=0.68-0.88; p<0.0001). A model of computed risk score [ Stage I-II (score 2)+Ki-67≥ 50% (score 1)+AR<10% (score 1)] predicted a probability of 67% for pCR when all three variables were favorable (Table). Conclusions: Four TIL groups were identified. In pts with moderate TIL levels, early stage disease, high Ki-67 and low AR were associated with increased probability of pCR with neoadjuvant therapy. [Table: see text]

Delineating longitudinal patterns of response to neoadjuvant systemic therapy (NAST) in triple-negative breast cancer (TNBC): Profiling results from a randomized, TNBC enrolling trial to confirm molecular profiling improves survival (ARTEMIS; NCT02276443)

586

Background: The heterogeneity of TNBC results in varied responses to NAST: 30-40% of patients (pts) have pathologic complete response (pCR) with excellent prognosis. Those with residual disease, have a much higher risk of recurrence. Longitudinal profiling assesses biologic response to NAST and mechanisms of resistance. Methods: Pts with stage I-III TNBC began a planned 4 cycles of Adriamycin-based chemo (AC). Biopsies were performed pre (mandatory) and post (optional) AC. Volumetric change by ultrasound (VUS) at completion of AC (or progression) was calculated. Pts with sensitive disease received subsequent taxane-based (T) therapy. Pts with insensitive disease were offered phase II trials. Pathologic response was assessed at surgical resection in 47 pts. Matched samples, pre and post AC (N = 48 pts) underwent transcriptomic and genomic profiling. Samples were classified into six previously identified ARTEMIS subtypes of TNBC (ART-Type). Immune deconvolution and estimation was performed using RNA-Seq profiles. Differential pathway-level analysis was performed comparing pre and post AC samples. Results: There was heterogeneity in response to AC with 4 predominate patterns of biologic response (Table). In 48% of cases the ART-Type of the tumor switched after AC, with androgen receptor like (LAR) and immune modulatory (IM) showing greatest stability. Tumors with enrichment in EMT or those with no significant dysregulation after AC (Groups C + D) were associated with less immune modulation and lower rates of pCR compared to those with depleted EMT (A and B) (8.7% vs 45.8%, p = 0.0078). Conclusions: Molecular profiling of longitudinal TNBC samples reveals distinct response patterns in tumors and their micro-environments upon treatment with AC. These patterns were indicative of pathologic response in this cohort; however, they require validation in a separate cohort. Clinical trial information: NCT02276443. [Table: see text]

A phase Ib study of entinostat plus lapatinib with or without trastuzumab in patients with HER2-positive metastatic breast cancer that progressed during trastuzumab treatment

BACKGROUND

Human epidermal growth factor 2 (HER2) is an effective therapeutic target in breast cancer; however, resistance to anti-HER2 agents such as trastuzumab and lapatinib develops. In a preclinical model, an HDAC inhibitor epigenetically reversed the resistance of cancer cells to trastuzumab and showed synergistic efficacy with lapatinib in inhibiting growth of trastuzumab-resistant HER2-positive (HER2+) breast cancer.

METHODS

A phase 1b, dose escalation study was performed to assess maximum tolerated dose, safety/toxicity, clinical efficacy and explored pharmacodynamic biomarkers of response to entinostat combined with lapatinib with or without trastuzumab.

RESULTS

The combination was safe. The MTD was lapatinib, 1000 mg daily; entinostat, 12 mg every other week; trastuzumab, 8 mg/kg followed by 6 mg/kg every 3 weeks. Adverse events included diarrhoea (89%), neutropenia (31%), and thrombocytopenia (23%). Neutropenia, thrombocytopenia and hypokalaemia were noted. Pharmacodynamic assessment did not yield conclusive results. Among 35 patients with evaluable response, PR was observed in 3 patients and CR in 3 patients, 1 maintained SD for over 6 months.

DISCUSSION

This study identified the MTD of the entinostat, lapatinib, and trastuzumab combination that provided acceptable tolerability and anti-tumour activity in heavily pre-treated patients with HER2+ metastatic breast cancer, supporting a confirmatory trial.