PDL-1 and immunohistochemistry markers in esophageal adenocarcinoma

e15567

Background: Esophageal adenocarcinoma (EAC) prognosis is poor and there is a need to identify patients that benefit most from neoadjuvant therapy. To examine the association of various biomarkers with clinical outcomes in neoadjuvant treatment of EAC, we retrospectively evaluated the biomarker expression (TS, ERCC1, TOPO1, PD-L1, PD-1) in patient matched formalin-fixed paraffin-embedded (FFPE) tumor samples. Methods: Immunohistochemistry of TS (TS106/4H4B1) , ERCC1 (Ab. 8F1), TOPO1 (1D6), PD-L1 (both 22c3 and SP142), PD-1 (NAT105), and chromogenic in-situ hybridization (CISH) of Her2 were performed on FFPE samples from 35 patients across 2 institutions at time of EAC diagnosis and after treatment when available. Retrospective clinical data and survival (5/2006-1/2016) was analyzed with a mean follow up of 110 months (range 22-306). Results: Overexpression (pre/post-treatment) of TS (60%/54%), ERCC1 (69%/16%), TOPO1 (74%/50%), PD-1 (54%/63%), PD-L1 (SP142) (2.9%/4%), PD-L1 (22c3) (0%/4%) and amplification of Her2 (18%/23%) were observed. Pretreatment observed PD-L1 levels were lower in our study (3%) when compared to other studies in EAC specimens (35%). Immunohistochemistry and changes observed after chemoradiation are reviewed in Table. No markers had significant correlation with prognosis however TS negative expression showed a non-significant (p=0.15) trend towards improved survival. Conclusions: Analyzing biomarkers in our neoadjuvant EAC cohort demonstrated a lower than expected PD-L1 positivity. In the largest cohort, to our knowledge, of patient matched FFPE tumor samples, we did not observe a statistically significant association between TS, ERCC1, TOPO1, PD-L1, or PD-1 with improved clinical outcomes. [Table: see text]

Phase I/II Randomized Trial of Sorafenib and Bevacizumab as First-Line Therapy in Patients with Locally Advanced or Metastatic Hepatocellular Carcinoma: North Central Cancer Treatment Group Trial N0745 (Alliance)

BACKGROUND

Angiogenesis has been a major target of novel drug development in hepatocellular carcinoma (HCC). It is hypothesized that the combination of two antiangiogenic agents, sorafenib and bevacizumab, will provide greater blockade of angiogenesis.

OBJECTIVE

To determine the optimal dose, safety, and effectiveness of dual anti-angiogenic therapy with sorafenib and bevacizumab in patients with advanced HCC.

PATIENTS AND METHODS

Patients with locally advanced or metastatic HCC not amenable for surgery or liver transplant were eligible. The phase I starting dose level was bevacizumab 1.25 mg/kg day 1 and 15 plus sorafenib 400 mg twice daily (BID) days 1-28. In the phase II portion, patients were randomized to receive bevacizumab and sorafenib at the maximum tolerated dose (MTD) or sorafenib 400 mg BID.

RESULTS

Seventen patients were enrolled in the phase I component. Dose-limiting toxicities included grade 3 hand/foot skin reaction, fatigue, hypertension, alanine/aspartate aminotransferase increase, dehydration, hypophosphatemia, creatinine increase, hypoglycemia, nausea/vomiting, and grade 4 hyponatremia. Seven patients were enrolled in the phase II component at the MTD: sorafenib 200 mg BID days 1-28 and bevacizumab 2.5 mg/kg every other week; 57% (4/7) had grade 3 AEs at least possibly related to treatment. No responses were observed in the phase II portion. Estimated median time to progression and survival were 8.6 months (95% CI: 0.4-16.3) and 13.3 months (95% CI 4.4 - not estimable), respectively.

CONCLUSIONS

The MTD of the combination is sorafenib 200 mg twice daily on days 1-28 plus bevacizumab 2.5 mg/kg on days 1 and 15 of a 28-day cycle. In the phase II portion of the trial, concerns regarding excessive toxicity, low efficacy, and slow enrollment led to discontinuation of the trial. (Clinical Trials ID: NCT00867321.).

Tumor mutational load in gynecological and breast cancer

44

Background: Lack of test standardization and conflicting results on PD1 and PD-L1 immunohistochemistry (IHC) assays challenge their use as predictive biomarkers for checkpoint inhibitors. High tumor mutational load (TML) has been linked to therapeutic response with immune checkpoint inhibition in melanoma, lung and colorectal cancer. Herein, we explore association of TML and PD1/PD-L1 expression in gynecological (GC) and breast cancer (BC). As secondary aim we explore the association of TML with BRCA1/2 mutations. Methods: De-identified data from molecular profiling on GC and BC tumors from the CARIS Life Sciences database was analyzed after obtaining IRB approval. TML was defined as the total number of nonsynonymous, somatic mutations per Mb sequenced with a 592-gene panel. PD-1 IHC was tested on Tumor Infiltrating Lymphocytes (TIL) with the cutoff of 1/HPF; PD-L1 expression was measured on tumor cells with the cutoff of 2+, 5%. Biomarker association was tested with ANOVA test (SPSS v23, IBM). Results: As shown in table, cervical (CC) and endometrial cancers (EC) had the highest TML and squamous CC and triple negative BC (TNBC) tumors with highest PD-L1. In EC, MSI-H was associated with higher TML when compared to MSS (20.1 vs. 6.1, p<0.001); 4 POLE-mutated tumors all carried TML (36-678) higher than mean (11.5). TML was not associated with PD1 or PDL1 positivity in any of the diseases. BRCA1/2 mutations were associated with higher mean TML in ovarian cancer (OC) (6.8 vs. 5.4 in wild type, p<0.001) but not in BC (6.3 vs. 6.8 in wild type, p=0.48). Conclusions: In this sample of patients who underwent molecular profiling, there was no association between TML and PD1 or PD-L1 expression. Association between BRCA1/2 tumor mutations and TML in OC is worth exploring. High TML in CC and EC support the development of immunotherapy for these diseases. [Table: see text]

Apocrine carcinoma of the breast: A brief update on the molecular features and targetable biomarkers

Apocrine carcinoma of the breast is a rare, primary breast cancer characterized by the apocrine morphology, estrogen receptor-negative and androgen receptor-positive profile with a frequent overexpression of Her-2/neu protein (~30%). Apart from the Her-2/neu target, advanced and/or metastatic apocrine carcinomas have limited treatment options. In this review, we briefly describe and discuss the molecular features and new theranostic biomarkers for this rare mammary malignancy. The importance of comprehensive profiling is highlighted due to synergistic and potentially antagonistic molecular events in the individual patients.

Landscape of Phosphatidylinositol-3-Kinase Pathway Alterations Across 19 784 Diverse Solid Tumors

Importance

Molecular aberrations in the phosphatidylinositol-3-kinase (PI3K) pathway drive tumorigenesis. Frequently co-occurring alterations in hormone receptors and/or human epidermal growth factor receptor 2 (HER2) may be relevant to mechanisms of response and resistance.

Objective

To identify patterns of aberration in the PI3K and interactive pathways that might lead to targeted therapy opportunities in clinical practice.

Design, Setting, and Participants

From January 2013 through December 2014, 19 784 consecutive tumor samples (>40 cancer types) were sent from thousands of clinicians in 60 countries to a single commercial laboratory for molecular profiling, including next generation sequencing, protein expression (immunohistochemical analysis [IHC]), and gene amplification (fluorescent in situ hybridization or chromogenic in situ hybridization).

Main Outcomes and Measures

Patterns in targetable genomic and proteomic alterations in the PI3K pathway and coincidence with hormone receptor and HER2 alterations.

Exposures

Molecular profiling across solid tumors.

Results

Overall, 38% of patients had an alteration in 1 or more PI3K pathway components, most commonly phosphatase and tensin homologue (PTEN) loss (by IHC) (30% of all patients), followed by mutations in PIK3CA (13%), PTEN (6%), or AKT1 (1%). Seventy percent of patients with endometrial cancer and more than 50% of patients with breast, prostate, anal, hepatocellular, colorectal, and cervical cancer exhibited alterations in at least 1 PI3K pathway gene and/or gene product. Examples of frequent aberrations included PTEN loss in hepatocellular (57% of patients), colorectal (48%), gastric (36%), prostate (52%), and endometrial cancer (49%); PIK3CA mutations in endometrial (37%), breast (31%), cervical (29%), and anal cancer (27%). PIK3CA, PTEN, and AKT1 mutations occurred more frequently in the presence of hormone receptor overexpression (androgen, progesterone, or estrogen receptor). PIK3CA mutations were also more common in the HER2-positive than in the HER2-negative group; the opposite pattern was seen for PTEN mutation or PTEN loss.

Conclusions and Relevance

PI3K pathway aberrations are among the most common in cancer. They do not segregate by classic cancer histologic characteristics. Patterns of biomarker coalterations involving HER2 and hormone receptors may be important for optimizing combination treatments across cancer types.

Abstract P2-05-09: Programmed death 1 (PD-1) and PD-1 ligand (PD-L1) distribution in triple negative breast cancer (TNBC)

Background: Recent data indicate a promising response to immune checkpoint inhibition in patients with metastatic TNBC. Ample research showed that PD-L1, a PD-1 ligand, is expressed in multiple tumor types, including TNBC, and may be a predictor of response to PD-1/PD-L1 blockade. Quantification of the stromal composition, particularly PD-1 and PD-L1 expression, continues to be controversial in its relationship to immune checkpoint inhibition in several cancer types, and it remains unclear whether PD-L1 expression is necessary to predict response. Here, we aimed to determine the distribution of PD-1 and PD-L1 in a large set of centrally ascertained specimens of TNBC.

Methods: The study cohort consisted of 993 tumor samples (both primary and metastatic TNBC) analyzed for either PD-1 or PD-L1 expression in one laboratory (Caris Life Sciences; Phoenix, AZ). Estrogen receptor and progesterone receptor status was assessed by immunohistochemistry (IHC). HER2/Neu expression or amplification was assessed by either IHC or in-situ hybridization. PD-1 and PD-L1 expression were confirmed using IHC with validated antibodies. For PD-L1, clone SP142 (Roche Diagnostics) was utilized and a sample was considered positive if there was &gt; 5% membranous staining of tumor cells. For PD-1, clone EH21.1 (BD Biosciences) was used. Tumor infiltrating lymphocytes (TILs) expressing PD-1 were counted and a sample was considered positive if there was at least one PD-1 positive TIL per 40x microscopic field.

Results: The median age in this cohort was 56 years (range: 22 – 88). A total of 363 TNBC specimens were tested for PD-1 via IHC. One hundred fifty eight (158; 43.5%) were negative for PD-1 expression. Two hundred five (205; 56.5%) were positive for PD-1. Of those that were PD-1 positive, 116 (56.6%), were in samples from a primary site (breast) and 89 (43.4%) in samples from a metastatic site. A total of 630 TNBC specimens were tested for PD-L1 via IHC. Five hundred seventy four (574; 91.1%) were negative for PD-L1. Fifty-six (56; 8.9%) were positive for PD-L1. Of those that were PD-L1 positive, were equally distributed between primary site and metastatic sites (28/324, and 28/306, respectively).

Conclusion: In this retrospective analysis, we describe, to the best of our knowledge, the distribution of PD-1 and PD-L1 expression in one of the largest datasets reported in TNBC. Unlike prior reports showing a high PDL-1 expression in excess of 50% in TNBC, this analysis show a low distribution of PD-L1 positivity. Our cohort represents a biased sample as those were unselected patients with recurrent breast cancer. Additionally, other factors can be implicated, including a change in the antibody used. These findings call for future standardization of the PD-L1 assay, particularly if further exploration showed PD-L1 to be a predictive or prognostic biomarker in mTNBC, particularly in relationship to therapy with immune checkpoint blockade.

Citation Format: Baker TM, Gatalica Z, Goldstein LJ, Obeid E. Programmed death 1 (PD-1) and PD-1 ligand (PD-L1) distribution in triple negative breast cancer (TNBC) [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P2-05-09.

Abstract P4-12-08: Use of an aptamer library based next generation omics platform for the development of a novel trastuzumab predictive assay

Introduction: Previous attempts to use individual aptamers as diagnostic reagents have failed to consistently achieve performance comparable to antibodies. Here we report a novel systems biology approach using poly-ligand aptamer libraries to identify responders and non-responders to traztuzumab-based regimens in metastatic breast cancer.

Methods: To overcome the fundamental limitation of the individual aptamer binding affinities, large libraries (106 species) were created so that potentially thousands of aptamers could bind to each of a multitude of targets related to the whole cellular changes in response to trastuzumab therapy. A set of breast cancer patients, which received trastuzumab mono- or combined therapy for at least 7 months were classified as “Responders” (R); cases with particular regimen discontinued in the period not exceeding 5 months were classified as “Non-Responders”(NR). A library of 2x1012 unique 90-mer ssDNA oligodeoxynucleotides (ssODN) was trained on FFPE tissue of both R and NR patients. Partitioning of aptamer libraries was done by microdissection of the tumor tissue, after incubation of aptamer library with the entire tissue section, to drive selection pressure toward cancer cells. A total of 10 cases of R and NR, 6 Her2+ cases each, were used to train separate aptamer libraries, with 1 positive and 2 counter selection cases per enrichment. Enriched libraries were screened on 20 R and 20 NR cases (11 Her2+ cases each) by adopting modified immunohistochemistry protocol. Each library was used as an independent reagent (similar to an antibody in IHC) across all 40 cases to evaluate the efficacy of the aptamer library to distinguish differences between the R and NR groups. Staining (DAB chromogen) profiles were scored from 0 to 3+ (nuclear and cytoplasmic staining) by a pathologist without any knowledge of the clinical outcomes. Initial validation was done by t-test using raw histological scores. Four libraries showed significant p-values between groups of responders and non-responders, a classification algorithm was constructed and evaluated using area under the receiver-operator characteristic curve (AUC). The datasets of two best-performing libraries were combined into one model using logistic regression to further improved the classifier performance.

Results: Of seventeen trained libraries, eight were evaluated and four showed significant correlation to clinical benefit with a minimum accuracy of 75% for each library when evaluated independently. Furthermore, two libraries showed exceptional performance (ROC curve AUC of 0.86 and 0.77). Combination of the profiling data from these two libraries using logistic regression resulted in an AUC of 0.985. A prospective validation of aptamer histochemical theranostic testing has been initiated.

Summary: Enriched aptamer libraries appear to distinguish trastuzumab responsiveness in metastatic breast cancer. This technology could be used as an additional technique beyond FISH testing to determine sensitivity to anti-HER2 agents. The demonstrated platform is applicable to virtually any disease where the safe and effective use of corresponding drug is yet to be improved.

Citation Format: Spetzler D, Domenyuk V, Santhanam R, Wei X, Stark A, Wang J, Gatalica Z, Miglarese M, Vidal G, Schwartzberg LS. Use of an aptamer library based next generation omics platform for the development of a novel trastuzumab predictive assay [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P4-12-08.

Abstract PD8-04: Racial differences in the molecular landscape of breast cancer

Background:African Americans (AA) have a higher mortality associated with breast cancer (BC) when compared to Caucasians (CC).This has been attributed to diverse factors that include access to care, reproductive factors and different somatic genomic profiles. We aimed to compare the racial mutational landscape of 565 BC samples.

Methods:DNA from formalin-fixed paraffin-embedded samples was sequenced using the Illumina NextSeq (Agilent SureSelect XT, 592 gene selected based on COSMIC database) and MiSeq (TruSeq, 47 gene) for mutation and gene amplification analyses. Protein expression was evaluated by Immunohistochemistry (IHC). The exome aggregation consortium database was assessed for known ethnicity associations.

Results:Tumor samples from 118 AA and 447 CC female patients were included in this analysis. AA were younger (median age 56 vs58y, p0.005) and had higher proportion of triple negative BC (TNBC) (32% vs17%, p0.001). 50.3% of the samples were obtained from primary tumors and the remainder from metastatic sites. This was similar in AA and CC (48.8% vs51.8% primary tumors, pNS). The two genes with highest mutation prevalence were TP53 and PIK3CA. AA had fewer PIK3CA mutations (14.7% vs28.2%, p0.03). Within HR+/HER2+ and HR+/HER2- subtypes there was a similar trend in the number of PIK3CA mutations but it was no longer significant. The remainder mutation analysis did not differ between races. In terms of protein expression there were significant differences in the androgen receptor(AR), RRM1, EGFR and TS expression (table). AR positivity defined as ≥10% was less frequent in AA (40.0% vs60.4%, p0.0001 and when adjusted for age, p0.005) and associated with PIK3CA mutations in both AA and CC (p0.01 and p0.007). AR expression in TNBC was positive in 17.8% of CC and 5.4% of AA (pNS). Copy number variation (CNV) data assessed by NextGen revealed significantly higher gene copy number in AA compared to CC in CCND1 (16% vs4%, p0.04), FGF19 (16% vs1.3%, p0.01) and FGF4 (16% vs3%, p0.02). When only TNBC was considered, RRM1, TOPO1 and TUBB3 expression was significantly higher in AA than CC (table) and there were no differences in the mutational analyses. Evaluation of other BC subtypes (HER2, HR positive) is currently underway.

Conclusions:In this large cohort of AA who underwent genomic profiling there were relatively few differences in the mutation analysis compared to CC. The only significant difference seen was the lower number of PIK3CA mutations in AA, which had been previously reported in a cohort of 105 AA from the TCGA data (Keenan et al. JCO 2015). Protein expression by IHC revealed lower expression of AR in AA, even after adjustment for age, which could have therapeutic implications. Some of the racial differences found in the molecular landscape of BC including PIK3CA mutations, AR, EGFR expression and CNV may contribute to a more aggressive tumor biology in AA.

Protein expression by IHCBCTNBCAA %CC %pAA %CC %pHER2/neu912ns00nsAR40600.005180.07ER44620.0000nsPR35460.0900nsEGFR30160.036050nsERCC133460.063626nsMGMT6163ns6571nsPD-150340.066548nsPD-L106ns011nsPGP76ns1015nsPTEN5658ns4247nsRRM139270.0459270.01TLE350590.104357nsTOP2A7164ns86470.08TOPO16358ns69460.03TS44300.016349nsTUBB355410.0677510.03

Citation Format: Lynce F, Xiu J, Nunes MR, Swain SM, Gatalica Z, Isaacs C, Pohlmann P. Racial differences in the molecular landscape of breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr PD8-04.

Abstract P6-07-27: Characterization of neuroendocrine breast carcinomas for biomarkers of therapeutic options

Introduction: Neuroendocrine breast carcinoma (NBC) is an extremely rare type of cancer, constituting less than 0.1% of all breast tumors, without specific treatment options. We investigated a biomarker database for frequency of molecular markers that may guide personalized treatment choices for these patients.

Materials and Methods: Molecular profiles of 40 breast carcinomas with neuroendocrine features [ER/PR+ (n=20), HER2+ (n=1) and TNBC (n=14)] were assessed (all female patients, mean age: 60.3 years, range: 39-83 years). Gene expression (Illumina DASL microarray platform), protein expression (IHC), gene amplification (ISH) and next-generation sequencing (NGS; TruSeq Illumina platform) were performed.

Results: 57% of NBCs were positive for hormone receptors (ER/PR), 40% were triple negative TNBC and 3% HER2+ subtypes. Therapeutic biomarkers (IHC) that may guide chemotherapies (and used in other primary sites neuroendocrine tumors) included: high TOP2A (85%) for etoposide or doxorubicin, low TS (57%) for 5-fluorouracil and low ERCC1 (45%) for cisplatin. Additional biomarkers for chemotherapy included: high TOPO1 (60%) for irinotecan, low RRM1 (48%) for gemcitabine and low MGMT (57%) for temozolomide. Biomarkers associated with available targeted therapies included: PTEN loss (39%), positive ALK (33%), cKit (30%), EGFR (29%), AR (26%) and PDGFRA (17%). No gene amplifications were detected in cMET, EGFR, or TOP2A. Targeted sequencing analysis of 47 genes detected variants in TP53, PIK3CA, ERBB4 and APC genes. Gene expression data (included somatostatin receptor gene family- SSTR1/2/3/4/5) was available for 5 patients, for which 3/5 patients exhibited overexpression of at least one SSTR gene.

Conclusions: Molecular profiling by a multiplatform approach reveals potential personalized therapy options for this very rare breast cancer subtype. With recent success of somatostatin analogs for other neuroendocrine tumors, the overexpression of SSTR gene family in NBC is worthy of further investigation.

Citation Format: Gatalica Z, Feldman R, Ghazalpour A, Vranic S. Characterization of neuroendocrine breast carcinomas for biomarkers of therapeutic options [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P6-07-27.

Molecular variances between rectal and left-sided colon cancers

522

Background: Recent analysis of CALGB 80405 showed that left sided colon tumors (LT) respond differently to biologics compared with right-sided tumors, likely due to molecular differences. Molecular variations between LT and rectal tumors remain undefined. Herein, we report our exploration of these variations. Methods: Tumors with origins clearly defined as splenic flexure to descending colon (SFT), sigmoid colon (SgT), or rectum (RT) were included. Protein expression, gene amplification and NextGen sequencing was tested. Microsatellite instability (MSI) was measured by PCR. Tumor mutational load (TML) was calculated using only somatic nonsynonymous missense mutations. Chi-square tests were used for comparative analyses. Results: In total, 1,457 primary tumors (SFT 125; SgT 460, and RT 872) were examined. When compared with SFT, RT had a higher frequency of TP53 (71% vs. 57%, p = 0.03) and APC (66% vs. 49%, p = 0.01); a lower frequency of PIK3CA (11% vs. 22%, p = 0.02), BRAF (3% vs. 15% p = 0.0001), GNAS (0.9% vs. 4%, p = 0.04), HNF1A (0.7% vs. 5%, p = 0.01), and CTNNB1(0.3% vs. 4%, p = 0.003); and a higher expression of TOPO1 (52% vs. 31%, p = 0.0001), ERCC1 (29% vs. 15%, p = 0.03), and MGMT (64% vs. 53%, p = 0.048). When compared with SgT, RT had higher expression of TLE3 (33% vs. 23%, p = 0.007), TOPO1 (52% vs. 35%, p < 0.001), TUBB3 (41% vs. 28%, p = 0.003), and MGMT (64% vs. 54%, p = 0.003). There were no differences between SFT, SgT, and RT in the frequency of PD-L1 expression (5%, 2%, and 2%) on tumor cells, PD-1 expression on tumor-infiltrating lymphocytes (54%, 42%, and 42%), or Her-2 expression (1%, 2%, and 3%) and amplification (3%, 3%, and 5%). MSI was seen in 7% of SFT, 4% of SgT, and 0.7% of RT (total LT vs. RT, p = 0.01). Mean TML was 23, 6.5, and 7 mutations (mut)/MB (332 tumors), and the portion of tumors carrying a TML of > 17mut/MB was 9%, 1.6%, and 4% for SFT, SgT, and RT, respectively. In all 3 cohorts, a TML > 17 mut/MB was highly concordant with MSI. There was a correlation between PD-1 and TML in RT (p = 0.04) but not in SFT or SgT. There were no correlations between PD-L1 and TML. Conclusions: Tumors arising in the rectum may carry genetic alterations that are distinct from LT. A better understanding of disease biology may help to identify therapeutic targets and advance precision medicine

Molecular characterization of squamous cell carcinoma of the anal canal (SCCA)

538

Background: Nivolumab has shown promising results in SCCA patients. The majority of SCCA cases have been linked to prior human papillomavirus (HPV) infection. However, HPV negative tumors are frequently TP53 mutated and often resistant to therapy. Molecular characteristics of SCCA are largely undefined. Here we explored the underlying biology of SCCA and the differences between TP53-wild type ( TP53-WT) and TP53-mutated ( TP53-MT) tumors. Methods: SCCA specimens underwent multiplatform testing with protein expression (IHC), gene amplification (ISH), and sequencing (NGS). Tumor mutational burden (TMB) was calculated using only somatic nonsynonymous missense mutations. Chi-square tests were used for comparative analyses. Results: In total, 253 tumors were studied. The most frequently mutated genes included PIK3CA (24%), BRCA2 (14%), FBXW7 (12.4%), TP53 (9.7%), and PTEN (8.9%). In a subset of 23 tumors subjected to Illumina NextSeq (592 gene) testing, the most common mutations were NOTCH2 (30%), NOTCH1 (27.3%), POLE (21.7%) , TSC2 (17.4%), PTEN (14.3%), BRAF (13.6%), BRCA2 (13.0%), PIK3CA (13.0%), and FBXW7 (9.5%). Tumors frequently expressed MRP1 (97.6%), EGFR (92.7%), TOP2A (88.5%), TOPO1 (69.5%), MGMT (67.8%), and RRM1 (59.9%). Expression of PD-1 was seen in 55.8% (24/43) of tumors, and PD-L1 in 15.4% (9/34). HER2 was amplified in 2% (3/147) of samples, which has not been previously described in SCCA. When compared with TP53-WT (n = 93) tumors, T P53-MT (n = 10) had higher rates of BRAF (22% vs. 1%, p < 0.001) and RB1 mutations (44% vs. 0%, p < 0.001), whereas TP53-WT had higher expression of TOPO1 (76% vs. 40%, p = 0.01) and TUBB3 (19% vs. 50%, p = 0.02). There were no differences between the two groups in the frequency of PD-1 or PD-L1 expression. Mean TMB was 8.6 mutations/megabase and, using a TMB cut-off > 17 mutations/megabase to define high vs. low TMB, 6.7% of tumors were TMB-High. High TMB did not correlate with PD-1 (p = 0.50) or PD-L1 status (p = 0.52). Conclusions: Molecular profiling differences between TP53-MT and TP53-WT SCCA indicate different carcinogenic pathways and biology, which may influence response to therapy. Low frequency mutations in several druggable genes may provide therapeutic opportunities for patients with SCCA.

Characterization of tumor mutation burden (TMB) in gastrointestinal (GI) cancers

530

Background: GI cancers are generally insensitive to immune checkpoint inhibitors (ICIs). Response to ICIs has been shown to correlate with TMB. Herein, we attempt to quantify TMB in GI cancers and its correlation with PD-1/PDL-1 expression. Methods: Tumor from various GI sites: right-sided and left-sided colon cancers (RT and LT), rectal cancer (RC), small bowel adenocarcinoma (SBA), gastric adenocarcinoma (GA), anal cancer (SCCA), hepatocellular carcinoma (HCC), esophageal adenocarcinoma and squamous cell carcinoma (EA and E-SCC), biliary cancer (BC), pancreatic adenocarcinoma (PA), and pancreatic neuroendocrine tumors (PNET), were analyzed by NextGen sequencing. TMB was calculated using only somatic nonsynonymous missense mutations sequenced with a 592-gene panel. MSI was assessed by fragment analysis. Correlation of PD-1/PD-L1 expression with TMB was calculated by student’s t test. Results: In total, 1375 tumors were examined. Among the different GI cancer types, RT and LT had the highest TMB (mean: 11.6 and 9.9 mutations [mut]/megabase [MB]), whereas BC and PA had the lowest levels (mean: 5.7 and 4.9 mut/MB) (Table). Overall primary tumors had higher TMB than metastases (mean: 8.3 vs. 6.5 mut/MB, p = 0.037). Using a cut-off of 17 mut/MB to define high vs. low TMB, high TMB was seen in all 24 MSI-H and 2 MSS colon tumors with POLE mutations, but not in other MSS colon tumors (n = 325, p < 0.0001). Similarly, among 6 GA tumors tested for MSI, high TMB was seen in 2 MSI-H while low TMB was seen in the 4 MSS tumors. Overall high TMB was seen most frequently in RT (12%), GA (11%), and SCCA (8%), and least frequently in PA (1.3%) and E-SCC (0%). PD-1 correlated with TMB in some tumor types (RT and RC), as did PD-L1 (RT and HCC). Conclusions: TMB varies among GI cancers. Forthcoming prognostic analysis to assess the correlation between TMB and response to ICIs in GI cancers is underway. [Table: see text]

Multiplatform molecular profiling identifies potentially targetable biomarkers in malignant phyllodes tumors of the breast

Malignant phyllodes tumor is a rare breast malignancy with sarcomatous overgrowth and with limited effective treatment options for recurrent and metastatic cases. Recent clinical trials indicated a potential for anti-angiogenic, anti-EGFR and immunotherapeutic approaches for patients with sarcomas, which led us to investigate these and other targetable pathways in malignant phyllodes tumor of the breast. Thirty-six malignant phyllodes tumors (including 8 metastatic tumors with two cases having matched primary and metastatic tumors) were profiled using gene sequencing, gene copy number analysis, whole genome expression, and protein expression. Whole genome expression analysis demonstrated consistent over-expression of genes involved in angiogenesis including VEGFA, Angiopoietin-2, VCAM1, PDGFRA, and PTTG1. EGFR protein overexpression was observed in 26/27 (96%) of cases with amplification of the EGFR gene in 8/24 (33%) cases. Two EGFR mutations were identified including EGFRvIII and a presumed pathogenic V774M mutation, respectively. The most common pathogenic mutations included TP53 (50%) and PIK3CA (15%). Cases with matched primary and metastatic tumors harbored identical mutations in both sites (PIK3CA/KRAS and RB1 gene mutations, respectively). Tumor expression of PD-L1 immunoregulatory protein was observed in 3/22 (14%) of cases. Overexpression of molecular biomarkers of increased angiogenesis, EGFR and immune checkpoints provides novel targeted therapy options in malignant phyllodes tumors of the breast.

Molecular profiling of 6,892 colorectal cancer samples suggests different possible treatment options specific to metastatic sites

Metastatic colorectal cancer (mCRC) carries a poor prognosis with an overall 5-year survival of 13.1%. Therapies guided by tumor profiling have suggested benefit in advanced cancer. We used a multiplatform molecular profiling (MP) approach to identify key molecular changes that may provide therapeutic options not typically considered in mCRC. We evaluated 6892 mCRC referred to Caris Life Sciences by MP including sequencing (Sanger/NGS), immunohistochemistry (IHC) and in-situ hybridization (ISH). mCRC metastases to liver, brain, ovary or lung (n = 1507) showed differential expression of markers including high protein expression of TOPO1 (52%) and/or low RRM1 (57%), TS (71%) and MGMT (39%), suggesting possible benefit from irinotecan, gemcitabine, 5FU/capecitabine and temozolomide, respectively. Lung metastases harbored a higher Her2 protein expression than the primary colon tumors (4% vs. 1.8%, p = 0.028). Brain and lung metastases had higher KRAS mutations than other sites (65% vs 59% vs 47%, respectively, p = 0.07, <0.01), suggesting poor response to anti-EGFR therapies. BRAF-mutated CRC (n = 455) showed coincident high protein expression of RRM1 (56%), TS (53%) and low PDGFR (22%) as compared with BRAF wild-type tumors. KRAS-mutated mCRC had higher protein expression of c-MET (47% vs. 36%) and lower MGMT (56% vs. 63%), suggesting consideration of c-MET inhibitors and temozolomide. KRAS-mutated CRC had high TUBB3 (42% vs. 33%) and low Her2 by IHC (0.5%) and HER2 by FISH (3%, p <0.05). CRC primaries had a lower incidence of PIK3CA and BRAF mutations in rectal cancer versus colon cancer (10% and 3.3%, respectively). MP of 6892 CRCs identified significant differences between primary and metastatic sites and among BRAF/KRAS sub-types. Our findings are hypothesis generating and need to be examined in prospective studies. Specific therapies may be considered for different actionable targets in mCRC as revealed by MP.