Clinicogenomic characterization of inflammatory breast cancer

Background

Inflammatory breast cancer (IBC) is a rare and poorly characterized type of breast cancer with an aggressive clinical presentation. The biological mechanisms driving the IBC phenotype are relatively undefined-partially due to a lack of comprehensive, large-scale genomic studies and limited clinical cohorts.

Patients and Methods

A retrospective analysis of 2457 patients with metastatic breast cancer who underwent targeted tumor-only DNA-sequencing was performed at Dana-Farber Cancer Institute. Clinicopathologic, single nucleotide variant (SNV), copy number variant (CNV) and tumor mutational burden (TMB) comparisons were made between clinically confirmed IBC cases within a dedicated IBC center versus non-IBC cases.

Results

Clinicopathologic differences between IBC and non-IBC cases were consistent with prior reports-including IBC being associated with younger age at diagnosis, higher grade, and enrichment with hormone receptor (HR)-negative and HER2-positive tumors. The most frequent somatic alterations in IBC involved TP53 (72%), ERBB2 (32%), PIK3CA (24%), CCND1 (12%), MYC (9%), FGFR1 (8%) and GATA3 (8%). A multivariate logistic regression analysis revealed a significant enrichment in TP53 SNVs in IBC; particularly in HER2-positive and HR-positive disease which was associated with worse outcomes. Tumor mutational burden (TMB) did not differ substantially between IBC and non-IBC cases and a pathway analysis revealed an enrichment in NOTCH pathway alterations in HER2-positive disease.

Conclusion

Taken together, this study provides a comprehensive, clinically informed landscape of somatic alterations in a large cohort of patients with IBC. Our data support higher frequency of TP53 mutations and a potential enrichment in NOTCH pathway activation-but overall; a lack of major genomic differences. These results both reinforce the importance of TP53 alterations in IBC pathogenesis as well as their influence on clinical outcomes; but also suggest additional analyses beyond somatic DNA-level changes are warranted.

Abstract P5-04-06: Breast Imaging Recommendations for Females <40 Years of Age with ≥20% Lifetime Breast Cancer Risk: Practice Patterns at a Specialized Clinic

Background— There are limited data to guide breast cancer screening recommendations among females < 40 yrs of age with elevated lifetime breast cancer risk not driven by a known germline mutation. The American Cancer Society recommends initiating screening at age 30, while the National Comprehensive Cancer Network (NCCN) recommends 10 yrs younger than the youngest affected relative (YAR). Both support screening MRI in addition to annual mammogram (MMG). This study describes practice patterns related to screening imaging recommendations and patient (pt) follow-through in young females with ≥20% lifetime breast cancer cared for in a specialized clinic.

Methods— At the Brigham and Women’s Hospital high-risk breast clinic, specialized advanced practice providers, surgeons and oncologists perform risk assessment including use of the Tyrer-Cuzick (TC) risk model, and provide risk management recommendations. For this study, we identified pts age< 40 yrs with >20% lifetime breast cancer risk, no known genetic mutation or high-risk breast lesions, and ≥1 first or second-degree relatives (FDR or SDR) with breast cancer. We evaluated factors associated with recommendation for i) early screening initiation, defined as prior to age 40, and ii) use of supplemental imaging modalities.

Results—335 pts met study criteria: 20% were age< 30, 36% were 30-34, and 44% were 35-39. Mean lifetime risk by the TC model was 32% (SD: 10%). Early screening was recommended in 75%; these pts were more likely to have an affected FDR (71% vs. 48%, p< 0.001) and younger affected relatives (median age of YAR: 44 vs. 55, p< 0.001). Among pts whose YARs were age< 50, early screening was recommended in-line with NCCN guidelines for 99% of pts with FDRs< 50 vs. 80% of pts with only SDRs< 50 (p< 0.001). Among pts whose YARs were age≥50, early screening was recommended contrary to NCCN guidelines in 51%. Factors associated with an early screening recommendation in this subgroup were having received a prior MMG (62% recommended early screening vs. 33% with no prior MMG) as well as being older at time of risk discussion (median age 37 in early screening group vs. 34 in routine screening group) and having younger affected relatives (median age of YAR: 53 vs. 56) (all p≤0.01). Regarding use of supplemental imaging, 35% were recommended screening MMG alone, while 65% were also offered screening MRI or ultrasound (US). Factors most strongly associated with offering MRI/US included having heterogeneously or extremely dense breasts, normal BMI, greater extent of family history, younger affected relatives and higher TC scores (Table). All except extent of family history remained statistically significant in multivariable analysis. Among those offered supplemental MRI/US who were eligible to initiate screening, 48% had pursued MRI, 7% US +/- MRI, 27% MMG alone, and 18% had no screening imaging at a median follow-up of 17 months.

Conclusions— These data suggest that providers in our high-risk breast clinic are using nuanced clinical judgment related to screening recommendations in pts < 40 yrs with elevated lifetime risk. Those with affected FDRs at age< 50 were consistently recommended early screening initiation, while practice recommendations varied more for pts with only SDRs age< 50 or those with YAR age≥50, suggesting a need for consensus criteria as to when to initiate screening in these subgroups. Multiple factors impacted recommendations for screening MRI/US, most notably breast density.

Factors associated with offering supplemental screening with MRI/US

Citation Format: Alexandra Wehbe, Alison Laws, Fisher Katlin, Eshita Sharma, Marybeth Hans, Mary Graichen, Brittany Bychkovsky, Rochelle Scheib, Judy Garber, Lydia Pace, Tari King. Breast Imaging Recommendations for Females <40 Years of Age with ≥20% Lifetime Breast Cancer Risk: Practice Patterns at a Specialized Clinic [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-04-06.

Abstract PD7-07: Somatic alterations in primary tumors of patients (pts) with metastatic breast cancer (MBC) may predict likelihood of brain metastasis

Background: Despite advances in treatment options, outcomes remain poor for many pts with breast cancer brain metastases (BCBMs). Identifying genomic predictors of brain metastasis from primary tumors could lead to better stratification of pts at risk and drive the development of preventative strategies. The objective of this study was to describe the landscape of genomic alterations in primary tumors from pts with MBC who subsequently did or did not develop BCBMs.

Methods: We performed a case control study to identify somatic alterations in primary tumors associated with a higher incidence of brain metastases. We reviewed outcomes for 2562 unique MBC patients from a single institution who underwent targeted next-generation DNA sequencing of > 280 cancer-related genes (OncoPanel) from their tumor between July 1, 2013 and December 31, 2020. Pts were included in this analysis if they had at least 2 years of follow-up from date of metastatic diagnosis and OncoPanel testing on a primary breast tumor. We compared single nucleotide variants (oncogenic or likely oncogenic), copy number variation (amplification and deep deletions) and tumor mutation burden in the primary tumors of pts in this cohort. Copy number variation was corrected for Panel version and tumor purity. Wilcoxon rank sum test and Fisher exact test was used to compare genomic differences between groups. False discovery rate was used to correct for multiple hypothesis testing and q < 0.1 was considered significant

Results: A total of 369 pts were included in the final analytic cohort. Of these, 115 were diagnosed with brain mets (cases, BM group) and 224 were not (controls, nBM group). The BM group was enriched for patients with HER2-positive breast cancer (33 vs 12.5%), consistent with previous work. In the whole cohort, the most common and clinically significant somatic alterations (oncogenic single nucleotide variants or copy number high amplification or two copy deletion) are shown in Table 1. When adjusting for subtype there were no significantly enriched SNVs in BM vs nBM group. When adjusting for subtype, FGFR1 amplification was significantly enriched in hormone receptor positive HER2 negative (HR+ HER2-) patients with BM (log2 odds ratio 1.22, q < 0.1). Tumor mutation burden was not significantly different in primary tumors between the BM and nBM groups (median TMB 7.3 vs 6.1, Wilcoxon p = 0.08).

Pathway analysis combining all subtypes revealed that RTK_RAS pathway (log2 odds ratio 1.64, q value < 0.1) and TP53 pathway (log2 odds ratio 1.15, q value < 0.1) gene sets were significantly enriched in the BM group. When controlling for subtype, pathway analysis revealed that RTK_RAS pathway gene set was significantly enriched in HR+ HER2- BM group (log2 odds ratio 1.36 q < 0.1).

Conclusions: In this case control series of patients with metastatic breast cancer with or without brain metastases, we found that primary tumors that are enriched for somatic alterations in the RTK_RAS and TP53 pathway may be associated with higher risk of developing brain metastases. Further validation in larger cohorts is warranted.

Table 1. Frequency of somatic alterations in primary tumor by brain metastasis outcome.

Citation Format: Sheheryar Kabraji, Yvonne Y. Li, Melissa E. Hughes, Hersh V. Gupta, Lauren Buckley, Janet L. Files, Ayesha Mohammed-Abreu, Anne-Marie Feeney, Greg Kirkner, Ashka Patel, Ana C. Garrido-Castro, Romualdo Barroso-Sousa, Brittany Bychkovsky, Matthew Meyerson, Sara Tolaney, Deborah A. Dillon, Bruce Johnson, Eric Winer, Andrew Cherniack, Nancy U. Lin. Somatic alterations in primary tumors of patients (pts) with metastatic breast cancer (MBC) may predict likelihood of brain metastasis [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD7-07.

Abstract PD14-09: PD14-09 The effect of timing of TP53 genetic testing on treatment and outcomes among women with Li-Fraumeni syndrome and breast cancer

Background: Li-Fraumeni syndrome (LFS) is a pan-cancer predisposition syndrome caused by pathogenic germline TP53 variants. Breast cancer (BC) is the most prevalent tumor in women with LFS. The risk of secondary malignancies, including multiple primary BCs, other LFS-related cancers, radiation-induced sarcomas, and local recurrences are important clinical concerns in the LFS setting. The diagnosis of LFS may influence treatment decisions and outcomes. Methods: In this international multicenter study, we analyzed women with pathogenic or likely pathogenic germline TP53 variants and BC (DCIS or invasive breast carcinoma) diagnosed 2002-2022 from three retrospective LFS cohorts (Dana Farber Cancer Institute, USA; Institut Gustave Roussy, France; Hospital Sírio-Libanês, Brazil). We excluded carriers of TP53 unconfirmed possibly mosaic variants, carriers of a 2nd pathogenic variant in another BC susceptibility gene, and those with missing data related to timing of genetic testing (TGT) or date of 1st BC diagnosis (dx). The overall cohort was divided in two groups: genetic testing before or at 1st BC dx (group A) and those with testing ≥1 year after 1st BC dx (group B). In cases with synchronous bilateral BC, we included the tumor of higher risk of recurrence (invasive, higher stage, more aggressive tumor biology) and excluded the other. The chi-square test was used to measure the association between TGT and other categorical variables. Results: 209 patients (pts) met criteria for this analysis. The median age of 1st BC dx was 35 years (IQR, 31-42). BC was the 1st cancer dx in 87.5% of the pts. Among 1st breast tumors, 38 were DCIS, 147 were early-stage BC (61 I, 49 II, 37 III) and 7 stage IV (17 missing). There were no differences between groups A and B regarding staging at dx. Missense TP53 variants were the most common type of germline mutation (n=154, 73.6%), with 60.4% (n=93) in the DNA-binding domain and 38.9% (n=60) in the tetramerization domain. Median follow-up from 1st BC dx was 6 years (IQR, 3-10). 53.1% of pts (n=111) underwent TP53 germline testing only after 1st BC dx. Family history of BC < 50 and non-BC malignancy prior to or synchronous with 1st BC dx were not associated with TGT (p=0.3 and p=0.2, respectively). 35.4% of pts developed a second primary BC (25 ipsilateral; 49 contralateral). Among pts without synchronous bilateral BC or metastatic BC at dx, 97 pts underwent contralateral risk reducing mastectomy (CRRM), 56.7% (55/97) as part of treatment surgery for the 1st BC. CRRM uptake was associated with TGT (A 70.3% vs B 41.6%, p=0.001). Of 194 pts with detailed data on surgical treatment (1st BC), 146 underwent mastectomies and 48 breast conserving surgery (BCS). Group A had more mastectomies (79.5% vs 61.2%, p=0.001) and less radiation therapy (10.2% vs 45.9%, p< 0.001). Among the irradiated pts, 9.8% (n=5) developed sarcomas in the irradiated field. Thirty-eight pts had BC recurrence: 21 loco-regional (A 6 vs B 15, p< 0.05), mostly in-breast, and 17 distant relapses. There was a significant statistical association between TGT and type of BC surgery (p=0.001), radiation-therapy (p< 0.001), CRRM uptake (p=0.001) and local relapses (p< 0.05). Conclusion: This analysis of BC in our sizable cohort of LFS patients with treatment data confirms that, timing of genetic testing affects some treatment options and outcomes, including surgical procedures and use or avoidance of radiation. These decisions appear to influence the risk of local recurrence or additional primary BC and radiation-induced sarcoma. Recognition of germline TP53 variants in breast cancer patients as part of genetic testing at diagnosis appears to have implications for treatment options and outcomes.

Citation Format: Renata Sandoval, Michele Bottosso, Natalia Polidorio, Brittany Bychkovsky, Benjamin Verret, Alessandra Gennari, Sophie Hyman, Maria Isabel Achatz, Olivier Caron, Fabrice Andre, Judy Garber. PD14-09 The effect of timing of TP53 genetic testing on treatment and outcomes among women with Li-Fraumeni syndrome and breast cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD14-09.

Abstract HER2-05: HER2-05 Comprehensive genomic characterization of HER2-low breast cancer

Background: About half of all breast cancers exhibit low HER2 expression. Despite lack of ERBB2 amplification, HER2-low tumors respond to trastuzumab deruxtecan (T-DXd), leading to the NCCN recommendation of T-DXd both for patients with HER2+ and HER2-low metastatic breast cancer (MBC). It remains however unclear if HER2-low represents a distinct molecular entity, as compared to HER2-0 MBC. Here, we compare the genomic landscape of HER2-low versus HER2-0 breast cancers in a large, single institution cohort. Methods: We identified consecutive patients with MBC seen at Dana-Farber Cancer Institute between 07/2013 and 12/2020. Patients were included if they had HER2-negative MBC per ASCO/CAP Guidelines and had undergone next generation sequencing (NGS) testing with a targeted, tumor-only platform (OncoPanel). Based on the HER2 status of the specimen tested by NGS, patients were divided into 2 groups: (i) HER2-low if immunohistochemistry (IHC) 1+ or 2+ non-amplified, or (ii) HER2-0 if IHC 0. Mutations of interest detected on NGS were classified as oncogenic using the OncoKB tool and additional annotation. Genomic profiles of HER2-low and HER2-0 tumors were compared using Chi-Square and Kruskal-Wallis tests. To determine genomic event enrichment between the two HER2 groups, logistic regression models were used, accounting for background rate and estrogen receptor (ER) expression. ERBB2 copy counts were calculated for tumors with recorded histology-estimated purities and copy-number segmentation using a simple model of allelic gain/loss. Results: Among 1847 patients with HER2-negative MBC, 1043 underwent NGS testing on a HER2-low (n=489, 47%) or HER2-0 sample (n=554, 53%). Most samples were metastatic (71%, n=743) while 29% (n=300) were from primary tumors. 73% had ductal histology, 13% were lobular and 14% had mixed or other histology. ER expression was enriched among HER2-low vs. HER2-0 tumors (76% vs. 60%; p< 0.001). Focusing on the most commonly occurring genetic mutations, no major differences were observed in HER2-low vs. HER2-0 tumors, after correcting for ER status (Table 1). Among all mutational events, any mutation in MPL, CYLD, and MAP3K and oncogenic mutations in TP53 and NF1 were more common in HER2-0, while any mutation in MTOR, RAD21, DNMT3A, and PDGFRA were enriched in HER2-low patients, when controlling for ER status and background mutational rate (p< 0.05). However, no mutation reached significance after accounting for multiple hypothesis testing. Similarly, no deep deletion or high amplification CNV events reached significance for either group. Analysis of tumor mutational burden in HER2-low vs. HER-0 tumors revealed no significant differences (median: 7.26 muts/Mb vs. 7.60 muts/Mb, p=1.00), including when accounting for ER status. Finally, among tumors with sufficient tumor purity for ERBB2 copy count analysis (n=374 and 419 for HER2-low and HER2-0, respectively), HER2-low tumors had a significantly higher number of ERBB2 alleles as compared to HER2-0 (< 2 copies, 15.0% vs. 30.9%, 2 copies 67.4% vs. 60.5%, and >2 copies, 17.6% vs. 8.6%; p< 0.001 by Kruskal-Wallis). Conclusions: To our knowledge, this is the largest comprehensive genomic analysis of HER2-low MBC to date. In our cohort of patients with HER2-negative MBC, the genomic landscape of HER2-low and HER2-0 tumors did not differ significantly, apart from a higher number of ERBB2 alleles. These data further support the notion that HER2-low, as currently defined, is not a distinct molecular subtype of breast cancer.

Citation Format: Paolo Tarantino, Hersh V. Gupta, Melissa E. Hughes, Janet L. Files, Sarah Strauss, Gregory Kirkner, Anne-Marie Feeney, Yvonne Y. Li, Ana C. Garrido-Castro, Romualdo Barroso-Sousa, Brittany Bychkovsky, Laura MacConaill, Neal Lindeman, Bruce Johnson, Matthew Meyerson, Sheheryar Kabraji, Rinath Jeselsohn, Xintao Qiu, Rong Li, Henry W. Long, Eric Winer, Deborah A. Dillon, Giuseppe Curigliano, Andrew Cherniack, Sara Tolaney, Nancy U. Lin. HER2-05 Comprehensive genomic characterization of HER2-low breast cancer [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr HER2-05.

Abstract P5-14-06: Tumor Genomic Landscape in Older Women with Metastatic Breast Cancer (MBC)

Background. Patients (pts) who develop MBC at older ages are underrepresented in clinical trials, are less likely to be included in comprehensive biomarker characterization studies, and experience worse breast cancer-specific survival than their younger counterparts. Elucidating genomic underpinnings of MBC and possible therapeutic targets for older breast cancer patients are critical priorities. Methods. We identified pts age >70 years at MBC diagnosis and a younger cohort (ages 50-69; age < 50), who were treated for MBC at a single center and who had their metastatic (or if not available, the primary) tumor, assessed by a targeted, tumor-only next generation sequencing (NGS) platform (OncoPanel) between 2013-2020. The NGS panel included mutations, copy number variation, tumor mutational burden (TMB), and hypermutation (HM) status, with mutations classified as oncogenic using the OncoKB tool and additional annotation. Copy number events were selected as being “oncogenic” if a high amplification was called for an oncogene or a deep deletion for a tumor suppressor. We compared findings for older (age >70) vs. younger (age < 50 and ages 50-69) MBC pts using Chi-Square and Kruskal-Wallis tests. To determine genomic event enrichment, logistic regression (LR) models were used, controlling for age (continuous), background rate, and tumor subtype (those with unknown subtype [n=27] were excluded from models). False discovery rate (FDR) was used to correct for multiple hypothesis testing. Results. The final analytic cohort included 2,380 pts. The median age at MBC diagnosis was 54.1 years overall (range 18.5- 91.9) and 73.6 years for those age >70. A total of 137 metastatic and 76 primary tumors were sequenced in pts age >70; in those age < 70, 1383 metastatic and 784 primary tumors were sequenced (for age < 50 [n=857] and 50-69 [n=1310]). Older pts were more likely to present with HR+/HER2- tumors (70.9% v. 62.4% v. 52.4%), and less likely to present with HER2+ (9.4% v. 14.4% v. 22.8%) or triple-negative breast cancer (TNBC) (18.8% v. 21.9% vs. 24.0%) at MBC diagnosis (listed >70, 50-69, < 50; P=1e-7). Older pts had higher average TMB vs. younger pts (9.57 in pts > 70, 8.56 in ages 50-69, 7.34 in ages < 50; P=3.5e-5). This was due to older pts having a higher incidence of hypermutation status as defined as TMB >10: 26.3% in age >70, 23.2% in ages 50-69, 16.8% in age < 50. Using q=0.1 as the threshold of significance, the presence of CDH1, PIK3CA, MAP3K1, TET2, and AKT oncogenic mutations were also enriched in older pts, while the presence of oncogenic GATA3, BRCA2, and TP53 mutations, as well as any mutation in BRCA1 were enriched in younger pts (too few oncogenic BRCA1 mutations were present for accurate modeling). The frequency of oncogenic PIK3CA mutations in HR+/HER2- tumors was highest in the oldest pts (44.4% in pts age >70 v. 31.6% in age 50-69 v. 26.7% in age < 50). Of pts who had oncogenic BRCA1/2 mutations identified on tumor-only NGS testing and underwent clinical germline testing (n=7 v. 60 v. 67, oldest to youngest), older pts had the lowest incidence of germline BRCA pathogenic variants (14.3% vs. 47.2.% vs. 67.2%; p=0.01); most BRCA mutations identified on NGS testing in older patients were considered likely somatic. When assessing enrichment in copy number events, ERBB2, RAD21, and BRIP1 amplifications were all significantly less frequent in older pts (q< 0.1), even when accounting for tumor subtype. Conclusions. In a large cohort of pts with MBC, the mutational and copy number landscape for older pts differs from that in younger pts, even after controlling for tumor subtype. Key actionable findings include a higher proportion of high TMB and PIK3CA-mutated tumors, emphasizing the importance of genomic profile testing in this pt population and further exploration of efficacy and tolerability of relevant therapies in those age >70 years.

Citation Format: Hersh V. Gupta, Rachel Freedman, Melissa E. Hughes, Yvonne Y. Li, Gregory Kirkner, Janet L. Files, Sarah Strauss, Ana C. Garrido-Castro, Lauren Buckley, Romualdo Barroso-Sousa, Brittany Bychkovsky, Sara Tolaney, Laura MacConaill, Neal Lindeman, Bruce Johnson, Matthew Meyerson, Eric Winer, Deborah A. Dillon, Andrew Cherniack, Nancy U. Lin. Tumor Genomic Landscape in Older Women with Metastatic Breast Cancer (MBC) [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-14-06.

Abstract P5-02-18: HER2 status and response to neoadjuvant anti-HER2 treatment among patients with breast cancer and Li-Fraumeni syndrome

Background: Breast cancer (BC) is the most common tumor in women with Li-Fraumeni syndrome (LFS), with a cumulative incidence of 85% by the age of 60 years. However, LFS-related BC characteristics are still underexplored since most data derive from small retrospective cohorts. A variable enrichment in HER2-positivity (ranging from 34 to 80%) has been reported, but information regarding the response to anti-HER2 treatments are currently lacking. Moreover, data regarding the new emerging category of HER2-low are missing. Methods: Invasive BCs diagnosed in patients (pts) with TP53 germline pathogenic/likely pathogenic variant between 2002-2022 at Institut Gustave Roussy (France), Dana-Farber Cancer Institute (USA) and Hospital Sírio-Libanês (Brazil) were included. HER2 and hormone receptor (HR) expression were retrospectively retrieved from pathology records and evaluated according to ASCO/CAP recommendations in place at the time of diagnosis. HER2-positive cases were defined by an immunohistochemistry (IHC) score of 3+ and/or HER2 gene amplification by ISH; HER2-negative cases were classified as HER2-low (IHC 1+ or 2+ with negative ISH assay) or HER2-zero (IHC score 0). Pathologic complete response (pCR) was defined as ypT0/is and ypN0. Results: Among 197 invasive BCs identified in a total of 176 pts, 50.3% (n=99) were HER-positive. Among those, median age at BC diagnosis was 33 years (range 21-61) and the most frequent TP53 variants were missense mutations (n=68), affecting the DNA-binding domain in 70.6% of cases and the tetramerization domain in 29.4% of cases. Most BCs were invasive ductal carcinoma (n=90), with histologic grade 3 in 56.6% of cases. At diagnosis, most pts had early stage disease (34.3% stage I; 32.3% stage II; 21.2% stage III), while 6 pts presented de novo stage IV disease. Most tumors were HR-positive (76.8%, n=76), while 23.2% were HR-negative. 38 patients with HER2-positive BCs were treated with neoadjuvant therapy, 32 cases had post-neoadjuvant pathology reports available for pathological response classification. Among those, 26 (81.2%) were HR-positive and 6 (18.8%) HR-negative. Among pts with neoadjuvant treatment data, 87.1% received trastuzumab, which was combined with pertuzumab in 43.3% of cases; chemotherapy regimens included taxanes in all pts, anthracycline in 43.3% and platinum in 16.7%. 71.9% (n=23) of pts reached a pCR (69.2% among HR-positive and 83.3% among HR-negative), while 9 (28.1%) had residual disease; pCR rate was 82.4% among pts treated with an anthracycline-free regimen. At a median follow-up of 36 months, only one patient relapsed. Among HER2-negative BCs with available IHC score and ISH for HER2-low classification (n=85), 28 (32.9%) were HER2-low and 57 (67.1%) HER2-zero. Conclusions: In this first report of treatment results in BC pts with LFS, enrichment of HER2-positive BCs was confirmed and a remarkable pCR rate was observed with neoadjuvant treatment. Our findings require validation in a larger cohort, which is in progress. Collaborative efforts are essential for high quality data about BC treatment in this subgroup of pts.

Citation Format: Michele Bottosso, Renata Lazari Sandoval, Benjamin Verret, Natalia Polidorio, Olivier Caron, Alessandra Gennari, Brittany Bychkovsky, Sophie Hyman, Maria Isabel Achatz, Valentina Guarneri, Fabrice Andre, Judy Garbe. HER2 status and response to neoadjuvant anti-HER2 treatment among patients with breast cancer and Li-Fraumeni syndrome [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P5-02-18.

Abstract P3-04-03: The value of screening MRI in patients with high-risk breast lesions: An observational single-institution cohort study

Background: The value of screening MRI in women with a history of breast atypical hyperplasia (AH) or lobular carcinoma in situ (LCIS), collectively termed high-risk lesions (HRL), remains unclear. In our high-risk clinic, we recommend annual mammograms (MMG) and consideration of annual breast MRI, with an emphasis on shared decision-making between patients and providers regarding MRI. Here we report cancer detection rates and outcomes by screening strategy.Methods: Patients with personal history of an HRL evaluated in the Brigham and Women’s Hospital high-risk clinic between 2015-2020 were identified from a prospective database. We excluded those whose HRL upgraded to malignancy at excision, or with a known pathogenic variant in a breast cancer gene. Patient characteristics associated with MRI use were compared with univariable and multivariable logistic regression. Kaplan Meier methods, univariable and multivariable Cox proportional hazards models were used to compare breast cancer detection and biopsy rates during follow-up.Results: Among 914 eligible patients with an HRL, 716 (78%) with at least two clinical visits were included for analysis. Screening MRI was used in 147 (21%) patients; 54% had 1 MRI, 29% had 2 MRIs, and 17% had 3-5 MRIs. Those receiving MRI were significantly younger and more likely to be white, premenopausal, have extremely dense breasts on MMG, normal body mass index, a family member age<50 with breast cancer, LCIS (vs. AH), and higher lifetime breast cancer risk by the Tyrer Cuzick model (all p<0.02). Chemoprevention (CP) was used in 276 (39%) patients and was more common in those receiving MRI (48% vs. 36%, p<0.01). In multivariable analysis, age (OR 0.96, p=0.02), white race (OR 2.98, p<0.01), LCIS (OR 1.72, p=0.03) and CP use (OR 1.68, p=0.01) remained significantly associated with MRI use, with a trend seen for extremely dense breasts (OR 1.6, p=0.07).At a median follow-up of 21 months (IQR: 11-35), 12 (1.7%) patients developed breast cancer, 4 receiving screening MRI and 8 not receiving MRI. The 3-year rate of breast cancer detection was 2.6% (95%CI: 0.8-8.0%) in those with MRI and 2.0% (95%CI: 0.9-5.4%) in those without MRI (p=0.53). Adjusting for age, LCIS vs. AH and CP use, MRI screening was not associated with increased likelihood of cancer detection (HR 1.35, p=0.64). Patients having MRI screening were significantly more likely to require a biopsy during surveillance; 3-year biopsy rates were 24.0% vs. 8.5% (p<0.01).Of the 12 cancers, 3 were DCIS and 8 invasive carcinomas, all clinically node negative, hormone receptor positive and AJCC pathologic prognostic stage 0-I. The remaining 1 cancer was an angiosarcoma. Most of the invasive carcinomas were <2cm (7/8 [88%], median 1.2cm), grade 1-2 (7/8 [88%]), HER2 negative (7/8 [88%]) and pathologic node negative (5/7 [71%] with sentinel node biopsy performed). Of the 4 cancers in those receiving MRI, 50% were visible on MMG. The 2 mammographically occult lesions were 6mm and 2cm MR-detected masses, corresponding to a pT1aN0 invasive tubular carcinoma and a 1.5cm angiosarcoma, respectively.Conclusions: In our high-risk clinic, MRI use was selective, most commonly in women who were younger, white, had LCIS and/or extreme breast density. CP is routinely recommended, yet uptake rates were higher in those receiving MRI, suggesting patient interest in preventative health measures likely also played a role in MRI use. Among patients with an HRL, the 3-year cancer detection rate was 2%, with no difference in rates between those undergoing screening MRI vs. annual MMG alone. However, MRI was associated with a threefold increased risk of breast biopsy. All cancers detected were early stage and clinically node negative. While longer follow-up is needed, the value of MRI screening in patients with HRL appears to be low.

Citation Format: Alison Laws, Fisher Katlin, Marybeth Hans, Mary Graichen, Olga Kantor, Christina Minami, Brittany Bychkovsky, Lydia Pace, Rochelle Scheib, Judy Garber, Tari King. The value of screening MRI in patients with high-risk breast lesions: An observational single-institution cohort study [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 P3-04-03.

Genomic Characterization of de novo Metastatic Breast Cancer

PURPOSE

In contrast to recurrence after initial diagnosis of stage I-III breast cancer [recurrent metastatic breast cancer (rMBC)], de novo metastatic breast cancer (dnMBC) represents a unique setting to elucidate metastatic drivers in the absence of treatment selection. We present the genomic landscape of dnMBC and association with overall survival (OS).

EXPERIMENTAL DESIGN

Targeted DNA sequencing (OncoPanel) was prospectively performed on either primary or metastatic tumors from 926 patients (212 dnMBC and 714 rMBC). Single-nucleotide variants, copy-number variations, and tumor mutational burden (TMB) in treatment-naïve dnMBC primary tumors were compared with primary tumors in patients who ultimately developed rMBC, and correlated with OS across all dnMBC.

RESULTS

When comparing primary tumors by subtype, MYB amplification was enriched in triple-negative dnMBC versus rMBC (21.1% vs. 0%, P = 0.0005, q = 0.111). Mutations in KMTD2, SETD2, and PIK3CA were more prevalent, and TP53 and BRCA1 less prevalent, in primary HR⁺/HER2^- tumors of dnMBC versus rMBC, though not significant after multiple comparison adjustment. Alterations associated with shorter OS in dnMBC included TP53 (wild-type: 79.7 months; altered: 44.2 months; P = 0.008, q = 0.107), MYC (79.7 vs. 23.3 months; P = 0.0003, q = 0.011), and cell-cycle (122.7 vs. 54.9 months; P = 0.034, q = 0.245) pathway genes. High TMB correlated with better OS in triple-negative dnMBC (P = 0.041).

CONCLUSIONS

Genomic differences between treatment-naïve dnMBC and primary tumors of patients who developed rMBC may provide insight into mechanisms underlying metastatic potential and differential therapeutic sensitivity in dnMBC. Alterations associated with poor OS in dnMBC highlight the need for novel approaches to overcome potential intrinsic resistance to current treatments.

Tumor Mutational Burden and PTEN Alterations as Molecular Correlates of Response to PD-1/L1 Blockade in Metastatic Triple-Negative Breast Cancer

PURPOSE

Few patients with metastatic triple-negative breast cancer (mTNBC) benefit from immune checkpoint inhibitors (ICI). On the basis of immunotherapy response correlates in other cancers, we evaluated whether high tumor mutational burden (TMB) ≥10 nonsynonymous mutations/megabase and PTEN alterations, defined as nonsynonymous mutations or 1 or 2 copy deletions, were associated with clinical benefit to anti-PD-1/L1 therapy in mTNBC.

EXPERIMENTAL DESIGN

We identified patients with mTNBC, who consented to targeted DNA sequencing and were treated with ICIs on clinical trials between April 2014 and January 2019 at Dana-Farber Cancer Institute (Boston, MA). Objective response rate (ORR), progression-free survival (PFS), and overall survival (OS) were correlated with tumor genomic features.

RESULTS

Sixty-two women received anti-PD-1/L1 inhibitors alone (23%) or combined with targeted therapy (19%) or chemotherapy (58%). High TMB (18%) was associated with significantly longer PFS (12.5 vs. 3.7 months; P = 0.04), while PTEN alterations (29%) were associated with significantly lower ORR (6% vs. 48%; P = 0.01), shorter PFS (2.3 vs. 6.1 months; P = 0.01), and shorter OS (9.7 vs. 20.5 months; P = 0.02). Multivariate analyses confirmed that these associations were independent of performance status, prior lines of therapy, therapy regimen, and visceral metastases. The survival associations were additionally independent of PD-L1 in patients with known PD-L1 and were not found in mTNBC cohorts treated with chemotherapy (n = 90) and non-ICI regimens (n = 169).

CONCLUSIONS

Among patients with mTNBC treated with anti-PD-1/L1 therapies, high TMB and PTEN alterations were associated with longer and shorter survival, respectively. These observations warrant validation in larger datasets.

Abstract P5-12-02: PTEN alterations and tumor mutational burden (TMB) as potential predictors of resistance or response to immune checkpoint inhibitors (ICI) in metastatic triple-negative breast cancer (mTNBC)

Purpose: To date no biomarker has been identified that predicts response to ICI in mTNBC. This study aimed to explore if tumor genomic alterations correlate with efficacy of PD-1/PD-L1 inhibition in patients (pts) with mTNBC. Methods: Demographic, treatment response, and long-term outcome data were collected on patients with mTNBC treated at Dana-Farber Cancer Institute (DFCI) under several clinical trials incorporating PD-1/PD-L1 inhibitors, given as monotherapy or combined with chemotherapy (CT). Pts included in this analysis had available results of targeted exon sequencing performed using Oncopanel, our institutional gene sequencing panel, on archival tumor tissue. TMB was calculated by determining the number of non-synonymous somatic mutations that occur per megabase of exonic sequence data across all genes on the panel. High TMB was defined as 310 mutations/megabase. TMB and gene alterations were correlated with objective response rate (ORR) per RECIST 1.1, progression-free (PFS) and overall survival (OS). Results: A total of 50 pts with mTNBC were included in this analysis. At baseline, the median age was 55.9 years (31.8–75.9), 60% had ECOG 0 and 40% had ECOG 1, 72% had visceral metastasis, and 46% had received 31 prior lines of systemic therapy in the metastatic setting. While 26% of pts received monotherapy [pembrolizumab (n=7, NCT02447003); atezolizumab (n=6; NCT01375842)], 74% received combination with CT [pembrolizumab plus eribulin (n=31; NCT02513472); atezolizumab plus nab-paclitaxel (n=6; NCT01633970)]. PTEN alterations were present in 30% of pts (mutations = 7; one copy number loss = 7; two copy number loss = 1). Median follow-up was 14 months (1–40). Pts with tumors harboring PTEN alterations had lower ORR (7% vs 57%; P<0.001), shorter median PFS (2.3 vs 6.3 months; P=0.027), and shorter median OS (8.1 vs 20.1 months; P=0.012) compared to pts without PTEN alterations. The median TMB was 6.6 mut/Mb (1.2–50.8), and 23% of pts had a high TMB. While high TMB was not associated with higher ORR (P=0.56), it was associated with better median PFS (16.5 vs 2.4 months; P=0.017), and better median OS (not reached vs 13.5 months; P=0.026). Both PTEN status and TMB remained significantly associated with PFS in the multivariable model. Only PTEN status remained significantly associated with OS in the multivariable analysis with the same covariables. Ongoing analysis to better understand if these predictors are specific for predicting benefit to immunotherapy and/or a marker of chemotherapy resistance will be presented at the symposium. Conclusion: PTEN genomic alterations and TMB may impact benefit from PD-1/PD-L1 inhibitors largely administered with chemotherapy in mTNBC. These observations warrant prospective validation and may inform the importance of stratifying pts according to these characteristics in future randomized studies with ICI.

Table 1.Multivariable analysis for PFS Hazard ratioConfidence Intervalp-valueCombination therapy0.420.16 – 1.130.009Visceral metastasis1.310.63 – 2.770.46Previous lines of therapy1.020.09 – 0.700.85ECOG 12.11.06 – 1.280.034PTEN altered3.741.65 – 8.440.002Hypermutated tumors0.850.75 – 0.970.011

Citation Format: Barroso-Sousa R, Tyekucheva S, Pernas-Simon S, Exman P, Jain E, Garrido-Castro AC, Hughes M, Bychkovsky B, Di Lascio S, Umeton R, Files J, Lindeman NI, MacConaill LE, Hodi FS, Krop IE, Dillon D, Winer EP, Wagle N, Lin NU, Mittendorf EA, Tolaney SM. PTEN alterations and tumor mutational burden (TMB) as potential predictors of resistance or response to immune checkpoint inhibitors (ICI) in metastatic triple-negative breast cancer (mTNBC) [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-12-02.

Abstract P4-03-02: Casting a wide net: Finding actionable results in non-breast cancer (BC) genes on multi-gene panel testing (MGPT) in a BC cohort

Background: MGPT for hereditary cancer syndromes allows for concurrent analysis of genes associated with many different cancer types. This may lead to the identification of unexpected mutations in genes with no BC link. The objective of this study was to examine the landscape of pathogenic mutations in a BC cohort who underwent MGPT, to assess if there was clinical suspicion for identified mutations and if the results would affect subjects' medical management.

Methods: Retrospective review of subjects with BC seen at a single institution who underwent MGPT from 1/1/15- 5/31/18 was conducted. MGPT was defined as testing of more than the 9 genes associated with BC (ATM, BRCA1, BRCA2, CDH1, CHEK2, PALB2, PTEN, STK11, TP53). Deidentified pedigrees were analyzed by genetic counselors to determine whether there was clinical suspicion of the presence of the mutations using national testing guidelines or clinical diagnostic criteria.

Results: Among 3044 subjects, 365 (12%) were found to have one pathogenic mutation in at least one cancer susceptibility gene. Subjects with mutations in APC I307K, moderate-penetrance BC genes (NBN, RAD50, BARD1), and MUTYH were excluded from further analysis. We identified 52 pathogenic mutations in genes not typically associated with risk for BC in 51 (2%) subjects (table 1). There was clinical suspicion for the identified mutation in 17 (33%).

Table 1:Non-BC gene mutation landscape Number of MutationsClinical Suspicion (%)Lynch syndrome117 (64%)MLH110MSH221MSH632PMS254Ovarian181 (6%)BRIP1*111RAD51C40RAD51D30SHDx62 (33%)SDHA*30SDHC*32Other156 (40%)FH10HOXB13*32MITF32NF142VHL40CDKN2A21 (50%)Total5217 (33%)*Contains individuals that also have a mutation in a BC susceptibility gene

Conclusion: Of 3044 BC patients who underwent MGPT, 2% were found to have a pathogenic gene mutation that would have been missed by a smaller BC gene panel. Medical or surgical management would be affected by the MGPT result in 86% of subjects. Only 6% of subjects with genetic risk for ovarian cancer had a family history of this disease. The single FH and 3 of 4 VHL mutations are only associated with disease in the biallelic state; these findings do not affect the subjects' care, but have implications for reproductive risk. The HOXB13 mutations were found in female subjects only, but would have implications for their male relatives. NF1 mutations are associated with BC risk, but were included in this analysis due to a historically distinct clinical phenotype. Only 50% of NF1+ subjects had a clinical diagnosis or family history of NF1. In all cases, cascade testing was offered to at-risk family members, allowing for cancer and reproductive risk stratification and management. This study demonstrates how comprehensive MGPT can provide a more complete and personalized cancer risk assessment for BC patients and their families.

Citation Format: Culver S, Kipnis L, Stokes S, Bychkovsky B, Scheib R, Rana H, Garber J. Casting a wide net: Finding actionable results in non-breast cancer (BC) genes on multi-gene panel testing (MGPT) in a BC cohort [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-03-02.

Efficacy of Anti-HER2 Agents in Combination With Adjuvant or Neoadjuvant Chemotherapy for Early and Locally Advanced HER2-Positive Breast Cancer Patients: A Network Meta-Analysis

BACKGROUND

Several (neo)adjuvant treatments for patients with HER2-positive breast cancer have been compared in different randomized clinical trials. Since it is not feasible to conduct adequate pairwise comparative trials of all these therapeutic options, network meta-analysis offers an opportunity for more detailed inference for evidence-based therapy.

METHODS

Phase II/III randomized clinical trials comparing two or more different (neo)adjuvant treatments for HER2-positive breast cancer patients were included. Relative treatment effects were pooled in two separate network meta-analyses for overall survival (OS) and disease-free survival (DFS).

RESULTS

17 clinical trials met our eligibility criteria. Two different networks of trials were created based on the availability of the outcomes: OS network (15 trials: 37,837 patients); and DFS network (17 trials: 40,992 patients). Two studies-the ExteNET and the NeoSphere trials-were included only in this DFS network because OS data have not yet been reported. The concept of the dual anti-HER2 blockade proved to be the best option in terms of OS and DFS. Chemotherapy (CT) plus trastuzumab (T) and lapatinib (L) and CT + T + Pertuzumab (P) are probably the best treatment options in terms of OS, with 62.47% and 22.06%, respectively. In the DFS network, CT + T + Neratinib (N) was the best treatment option with 50.55%, followed by CT + T + P (26.59%) and CT + T + L (20.62%).

CONCLUSION

This network meta-analysis suggests that dual anti-HER2 blockade with trastuzumab plus either lapatinib or pertuzumab are probably the best treatment options in the (neo)adjuvant setting for HER2-positive breast cancer patients in terms of OS gain. Mature OS results are still expected for the Aphinity trial and for the sequential use of trastuzumab followed by neratinib, the treatment that showed the best performance in terms of DFS in our analysis.