TCTP regulates genotoxic stress and tumorigenicity via intercellular vesicular signaling

Oncogenic intercellular signaling is regulated by extracellular vesicles (EVs), but the underlying mechanisms remain mostly unclear. Since TCTP (translationally controlled tumor protein) is an EV component, we investigated whether it has a role in genotoxic stress signaling and malignant transformation. By generating a Tctp-inducible knockout mouse model (Tctp-/f-), we report that Tctp is required for genotoxic stress-induced apoptosis signaling via small EVs (sEVs). Human breast cancer cells knocked-down for TCTP show impaired spontaneous EV secretion, thereby reducing sEV-dependent malignant growth. Since Trp53-/- mice are prone to tumor formation, we derived tumor cells from Trp53-/-;Tctp-/f- double mutant mice and describe a drastic decrease in tumori-genicity with concomitant decrease in sEV secretion and content. Remarkably, Trp53-/-;Tctp-/f- mice show highly prolonged survival. Treatment of Trp53-/- mice with sertraline, which inhibits TCTP function, increases their survival. Mechanistically, TCTP binds DDX3, recruiting RNAs, including miRNAs, to sEVs. Our findings establish TCTP as an essential protagonist in the regulation of sEV-signaling in the context of apoptosis and tumorigenicity.

Most large structural variants in cancer genomes can be detected without long reads

Short-read sequencing is the workhorse of cancer genomics yet is thought to miss many structural variants (SVs), particularly large chromosomal alterations. To characterize missing SVs in short-read whole genomes, we analyzed 'loose ends'-local violations of mass balance between adjacent DNA segments. In the landscape of loose ends across 1,330 high-purity cancer whole genomes, most large (>10-kb) clonal SVs were fully resolved by short reads in the 87% of the human genome where copy number could be reliably measured. Some loose ends represent neotelomeres, which we propose as a hallmark of the alternative lengthening of telomeres phenotype. These pan-cancer findings were confirmed by long-molecule profiles of 38 breast cancer and melanoma cases. Our results indicate that aberrant homologous recombination is unlikely to drive the majority of large cancer SVs. Furthermore, analysis of mass balance in short-read whole genome data provides a surprisingly complete picture of cancer chromosomal structure.

Clinical Validation of Artificial Intelligence-Augmented Pathology Diagnosis Demonstrates Significant Gains in Diagnostic Accuracy in Prostate Cancer Detection

CONTEXT.—

Prostate cancer diagnosis rests on accurate assessment of tissue by a pathologist. The application of artificial intelligence (AI) to digitized whole slide images (WSIs) can aid pathologists in cancer diagnosis, but robust, diverse evidence in a simulated clinical setting is lacking.

OBJECTIVE.—

To compare the diagnostic accuracy of pathologists reading WSIs of prostatic biopsy specimens with and without AI assistance.

DESIGN.—

Eighteen pathologists, 2 of whom were genitourinary subspecialists, evaluated 610 prostate needle core biopsy WSIs prepared at 218 institutions, with the option for deferral. Two evaluations were performed sequentially for each WSI: initially without assistance, and immediately thereafter aided by Paige Prostate (PaPr), a deep learning-based system that provides a WSI-level binary classification of suspicious for cancer or benign and pinpoints the location that has the greatest probability of harboring cancer on suspicious WSIs. Pathologists' changes in sensitivity and specificity between the assisted and unassisted modalities were assessed, together with the impact of PaPr output on the assisted reads.

RESULTS.—

Using PaPr, pathologists improved their sensitivity and specificity across all histologic grades and tumor sizes. Accuracy gains on both benign and cancerous WSIs could be attributed to PaPr, which correctly classified 100% of the WSIs showing corrected diagnoses in the PaPr-assisted phase.

CONCLUSIONS.—

This study demonstrates the effectiveness and safety of an AI tool for pathologists in simulated diagnostic practice, bridging the gap between computational pathology research and its clinical application, and resulted in the first US Food and Drug Administration authorization of an AI system in pathology.

Abstract 1394: Comprehensive clinical and genomic analysis for patients with MYC, MYCN, and MYCL amplified solid tumors

Introduction: The MYC gene family (MYCf), which includes MYC, MYCN, and MYCL, is deregulated in ~70% of cancers and is associated with treatment resistance. Whereas older investigational therapies for MYC amplified tumors were unsuccessful, promising novel targeted therapies are in early phase clinical trials. Unfortunately, it remains unclear how to select patients whose cancers may harbor true MYC addiction. We thus sought to characterize factors such as amplification level, focality, and clonality that may correlate with increased MYC dependence.

Methods: Utilizing a center-wide next generation sequencing (NGS) program of >71,000 sequenced patients, genomic and clinical data from pediatric and adult patients with MYC, MYCN, and MYCL amplifications were identified between 2014 and 2022. Patients were characterized as harboring MYC, MYCN, and MYCL amplification based on a read-depth methodology using a DNA-based hybrid-capture NGS (MSK-IMPACT) and Fraction and Allele-Specific Copy Number Estimates from the Tumor Sequencing (FACETS). All cases underwent clinical data curation including baseline demographic, tumor characteristics, and treatment histories.

Results: We identified 3911 cancers with MYCf amplification (n=3257 (82%) MYC; n=364 (9%) MYCL; n=330 (8%) MYCN) across 40 malignancies, for an overall 5.5% incidence. The most frequent tumor types with MYCf amplification were breast (22%), non-small cell lung (NSCLC) (11%), colorectal (8%), ovarian (8%), prostate (7%), brain (5%), and small cell lung cancers (SCLC) (2%). Cancers with MYC amplification had longer segment lengths than MYCL and MYCN amplification, which appeared more focal (median = 19, 4.3 and 4.5 MB, respectively, p < 0.001). MYCN amplified cancers had higher total copy number than MYC and MYCL amplified cancers (median = 19, 8, 9, respectively, p < 0.001). MYC, MYCN, and MYCL samples were predominantly clonal (median clonal fraction > 99% for all genes). Most NSCLC, squamous cell lung cancers, and pulmonary carcinoids had MYC amplifications (93%, 70%, and 67% respectively). Conversely, SCLCs most often had MYCL amplifications (49%). No concurrent targetable driver alterations were found in 33% of metastatic NSCLCs with MYC, 75% of MYCN, and 6% of MYCL amplifications.

Conclusions: While MYCf amplification is observed across a broad range of cancer types, factors such as gene type (MYC, MYCN, MYCL), focality, total copy number, clonality, and concurrent oncogenic drivers vary widely. Novel MYC-directed trials may consider enrichment for a subpopulation of cancers with higher-level, focal, and clonal MYCf amplifications without concurrent other drivers.

Citation Format: Monica F. Chen, Allison Richards, Patrick Evans, Patrick Lee, Adam Price, Matteo Repetto, Soo Ryum Yang, Jason Chang, Rose Brannon, Ezra Rosen, David Brown, Charles Rudin, Nitya Raj, Mark G. Kris, Jorge Reis-Filho, Mark Donoghue, Alexander E. Drilon, Noura J. Choudhury. Comprehensive clinical and genomic analysis for patients with MYC, MYCN, and MYCL amplified solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1394.

Abstract 4284: Enhancing subclonal reconstruction algorithm for resolving complex tumor phylogenies from multi-sample tumor DNA sequencing

Cancer is characterized by the ongoing accumulation of somatic mutations, providing selective advantages that may lead to dysregulated cellular proliferation. While the cancer genome at diagnosis has been extensively studied, many cancer types still lack strong prognostic biomarkers. The continuous acquisition and selection for driver mutations in a population of cancer cells acts as a Darwinian process, resulting in clonal expansions of progressively more aberrant and fit phenotypes. Reconstructing tumor evolution allows us to understand key events that drive cancer progression and patterns of mutation co-occurrence within clones. These evolutionary features guide our understanding of fundamental mechanisms that lead to disease lethality. Inferring tumor evolution from DNA sequencing data is becoming part of routine analysis in cancer research. As sequencing costs drop, sequencing multiple tumor samples from a patient becomes routine. These multiple samples can represent different spatial regions of a tumor, longitudinal samples from a single region or a combination of both. This provides an opportunity to study tumor evolution in much greater detail and accuracy than was previously feasible through single-sample datasets. The most widely used methods to reconstruct the subclonal evolution of a tumor utilize stochastic-search algorithms. These approaches iterate through a parameter space to select phylogenetic solutions that maximize the likelihood of observed sequencing data. They are optimized for low complexity cases, where the size and number or subclones are relatively limited. As tumor subclonal structure increases in complexity, the parameter space grows exponentially, and stochastic-search algorithms become computationally intractable. For instance, recent benchmarking studies have revealed that many methods fail to reconstruct clone trees for data with as few as ten subclones. To circumvent current computational limitations, we developed a deterministic algorithm for subclonal reconstruction that leverages fundamental principles of cancer biology to encode heuristics that reduce the solution space to biologically plausible phylogenies. When applied to samples (4-36 tumors; median 16) from 12 patients with metastatic breast cancer, our method reduced the average runtime ten-fold. We were able to delineate the evolutionary history of up to 57 distinct subclones per patient, which is infeasible with most current methods. Benchmarking using methods developed for the SMCHet DREAM challenge on real and simulated datasets further quantifies the accuracy, resolution, and scalability. We have thus presented a novel method for rapid and optimized reconstruction of tumor evolutionary histories.

Citation Format: Helena Winata, Daniel Knight, Juber A. Patel, Nicholas K. Wang, Pier Selenica, Stefan E. Eng, Caroline Kostrzewa, Jaron Arbet, Yingjie Zhu, Ronglai Shen, Jorge Reis-Filho, Pedram Razafi, Paul C. Boutros. Enhancing subclonal reconstruction algorithm for resolving complex tumor phylogenies from multi-sample tumor DNA sequencing. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4284.

Abstract P2-23-15: Histologic, immunohistochemical and genomic comparison between classic Invasive lobular carcinomas and lobular-like invasive ductal carcinomas

Background: Invasive lobular carcinomas (ILCs) are the most frequent special histologic subtype of breast cancer, accounting for up to 15% of all breast cancer cases. ILCs are characterized by a distinctive discohesive growth pattern, with cells arranged in single cell infiltrative file and dispersed throughout the stroma, which stems from the loss of E-cadherin expression due to bi-allelic inactivation of the CDH1 gene. A subset of breast cancers display a similar single cell infiltrative growth pattern but, in contrast to classic ILC, display diffuse strong membranous E-cadherin reactivity and membranous p120 expression. We refer to such cases as “lobular-like invasive ductal carcinoma” (LLIDC), but it is unclear if this terminology is appropriate and if such cases show biallelic inactivation of CDH1, similarly to ILCs. Here, we sought to define whether LLIDCs would harbor bi-allelic alterations of CDH1 and to perform an exploratory, hypothesis generating analysis of the repertoire of somatic genetic alterations of LLIDCs and classic ILCs. Materials and methods: Representative H&Es, as well as sections subjected to E-cadherin and p120 immunohistochemistry from seven classic ILCs and seven bona fide “lobular-like invasive ductal carcinomas” were retrieved and independently reviewed by two pathologists with experience and expertise in breast pathology. DNA samples were extracted from representative sections from tumor and normal breast tissue from each patient and subjected to an FDA-approved targeted sequencing assay comprising the coding regions and selected regulatory elements of 515 genes. Somatic single nucleotide variants (SNVs) were detected with MuTect, indels with Strelka, Varscan2, Scalpel and Lancet. All mutations were manually inspected using the Integrative Genomics Viewer (IGV). The cancer cell fraction (CCF) of each mutation was inferred, as well as clonal probability, using ABSOLUTE. Copy number alterations and loss of heterozygosity were determined using FACETS. Mutational signatures were inferred using SigMA based on all synonymous and nonsynonymous somatic mutations. Results: Based on the histopathologic evaluation, of the 14 cases analyzed, seven were classified as ILC, and the other seven were classified as LLIDC. Sequencing analysis revealed that the classic ILCs harbored 16q LOH and CDH1 mutations (7/7), of which five were frameshift indel and two were splice site mutations consistently coupled with loss-of-heterozygosity (LOH) of the wild-type allele. Conversely, five of the seven LLIDCs did not harbor CDH1 mutations or genomic rearrangements. CDH1 mutations were identified in 2 LLIDCs: one harbored a subclonal CDH1 in-frame indel mutation coupled with LOH. This case displayed membranous E-cadherin and p120 expression with areas of aberrant expression. The other CDH1-mutated LLIDC harbored a complex in-frame indel with subclonal LOH. This case displayed membranous E-cadherin and p120 expression. The comparative analysis of the repertoire of somatic genetic alterations and mutational signatures present in LLIDCs and classic ILCs did not reveal any significant differences. Conclusion: Despite the histologic similarities, LLIDCs differ from classic lobular carcinomas based on the lack of CDH1 bi-allelic inactivation and the patterns of expression of E-cadherin and p120 catenin. Further whole-genome sequencing analyses are warranted to define the molecular basis of the discohesive cancer cells of LLIDC display.

Citation Format: Edaise M. da Silva, Thais Basili, Jing Yu, Juan Blanco-Heredia, Pier Selenica, Qiqi Ye, Arnaud da Cruz Paula, Higinio Dopeso, Antonio Marra, Steffi Oesterreich, Jorge Reis-Filho, Rohit Bhargava. Histologic, immunohistochemical and genomic comparison between classic Invasive lobular carcinomas and lobular-like invasive ductal carcinomas [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 P2-23-15.

Abstract P6-05-10: An international survey on invasive lobular breast cancer (ILC) reveals gaps in knowledge and top priority research areas

Background: There is growing awareness of the unique etiology, biology, clinical presentation and progression of Invasive lobular breast cancer (ILC), but additional research is needed to assure translation of findings into management and treatment guidelines. We performed a survey to: 1) analyze the landscape of the current understanding of ILC, and 2) identify consensus research questions on ILC.

Methods: The IRB-approved survey was developed with input from representatives of three major stakeholder groups - breast cancer clinicians/researchers, laboratory-based researchers, and advocates/patients. We fielded the survey from March to May 2022 using targeted email and via social media.

Results: 1,774 participants answered at least one question and 1,310 finished the survey. Participants are from 66 countries from all continents (except Antarctica). Respondents self-identified as clinicians (mostly medical oncologists and surgeons) (N=413), researchers (N=376), and breast cancer patients (1,121), with some belonging to more than one category. 26% of the patients who participated in the survey belong to advocate groups.

Only 46% of clinicians reported being confident in describing the differences between ILC and no special type (NST) (invasive ductal) breast cancer. Knowledge of histology was seen as important (73%), affecting their treatment decisions (51%), and refined treatment guidelines would be valuable for patients with ILC in the future (76%). 85% of clinicians have never powered a clinical trial to allow subset analysis for histological subtypes, but the majority would consider it. 88% would participate in a consortium to conduct clinical trials on ILC. The top two most important research questions were: 1) determining mechanisms of endocrine resistance, and, 2) identifying novel therapeutic targets, repurposing existing drugs and progressing them to clinical trials.

Of the researchers, 48% reported being confident in describing differences between ILC and NST. They reported that ILCs are inadequately presented in large genomic data sets (52%), and that ILC models are insufficient (42%). Only 13% of respondents have inadequate access to tissue or blood from patients with ILC. The top two most important research questions identified by the laboratory researchers overlapped with those identified by the clinicians, i.e. understanding of endocrine resistance and identifying novel drugs that can be tested in clinical trials.

The majority of patients (52%) thought that their health care providers did not explain unique features of ILC, and that in general communication was limited. When asked about top research question, they chose: 1) Improvement of ILC screening/early detection, and, 2) Identifying new and specific imaging tools for ILC.

When comparing top priority topics across six research domains, there was a high degree of consistency, especially among clinicians and researcher, but less so when compared with the breast cancer patients (Table 1).

Conclusion: In summary, we have gathered timely and representative information from an international community of clinicians, researchers, and patients/advocates that we expect will lay the foundation for a community-informed collaborative research agenda, with the goal of improving the management and personalizing treatment for patients with ILC.

Table 1. Ratings by all three stakeholder groups of the most critical and impactful ILC research topics. Top box scores between stakeholder groups were compared using chi-square analysis.

Citation Format: Steffi Oesterreich, Leigh Pate, ADRIAN V. LEE, Rachel C. Jankowitz, Patrick Derksen, Rita Mukhtar, Otto Metzger, Matthew J. Sikora, Christopher Li, Christos Sotiriou, Gary Ulaner, Jorge Reis-Filho, Nancy E Davidson, Karen Van Baelen, Laurie Hutcheson, Siobhan Freeney, Flora Migyanka, Claire Turner, Todd Bear, Christine Desmedt. An international survey on invasive lobular breast cancer (ILC) reveals gaps in knowledge and top priority research areas [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 P6-05-10.

Abstract HER2-07: HER2-07 Genomic Characterization of Primary and Metastatic HER2-low Breast Cancers

Background: Varying levels of HER2 expression without ERBB2 gene amplification can be detected by immunohistochemistry (IHC) in approximately 60% of all invasive breast cancers (BCs). HER2-low-expressing BCs have recently been shown to respond to novel anti-HER2 antibody drug conjugates. Several studies have demonstrated that HER2-low BCs do not seem to constitute a distinct clinical and transcriptomic subtype as compared to HER2-0 BCs. Here we sought to define the clinicopathologic features and repertoire of somatic genetic alterations in HER2-low BCs. Methods: We retrieved clinical, pathological, and genomic data of BCs that were subjected to targeted sequencing using the FDA-cleared MSK-IMPACT assay from April 2014 to December 2021. After removing cases where any available biopsy had HER2 3+ and/or positive ERBB2 fluorescence in situ hybridization (FISH), 3608 samples (primary=1347; post-treatment/metastatic=2261) were included. Tumors were classified as HER2-low if they had an HER2 IHC score of 1+ or 2+ with a non-amplified FISH assay and HER2-0 if they had an HER2 IHC score of 0. Somatic mutations and DNA copy number alterations from MSK-IMPACT were analyzed. Multiple testing correction using the Benjamini-Hochberg method was applied to control for the false discovery rate (q). Q values < 0.1 were considered significant. Results: Among 3608 HER2- BCs, 1460 (40%) and 2148 (60%) were HER2-0 and HER2-low, respectively. Hormone receptor (HR) expression was significantly higher in HER2-low than HER2-0 tumors in both primary (781 [68.3%] vs 362 [31.7%]; p< 0.001) and metastatic (1031 [60.5%] vs 673 [39.5%]; p< 0.001) samples. A higher proportion of HER2-low tumors was found in metastatic than primary samples (59% vs 41%; p< 0.001) in this cohort. No difference in histology subtype, tumor grade, disease stage (among primary tumors), mutational signatures, and tumor mutational burden was found overall and when cases were stratified by HR expression. In HR-positive BCs, HER2-0 BCs harbored higher frequency of TP53 (33% vs 25%; odds ratio [OR] 1.49, 95% confidence interval [CI] 1.25-1.78, q< 0.001) and CDKN1A (1% vs 0%; OR 17.47, 95% CI 2.48-756.37, q=0.02) alterations than HER2-low BCs. Similar findings were observed in metastatic but not in primary HR-positive BCs. No differences were detected in HR-negative BCs stratified into HER2-low and HER2-0. Given the potential misclassification that exists between IHC HER2-0 and HER2-1+, we then conducted an exploratory analysis splitting the HER2-low group into HER2 1+ and 2+. TP53 alterations remained significantly enriched in HER2-0 compared to HER2-1+ HR-positive tumors (33% vs 24%; OR 1.55, 95% CI 1.28-1.87, q< 0.001). In HR-positive BCs, HER2-2+ displayed a higher frequency of genetic alterations in genes encoding for transcription factors, such as MYC (14.2% vs 7.3%; OR 2.09, 95% CI 1.44-3.04, q=0.02) and YAP1 (2% vs 0.3%; OR 6.86, 95% CI 1.7-39.57, q=0.1), and DNA damage response, such as FAM175A (1.6% vs 0%; OR 18.23, 95% CI 2.43-807.73, q=0.03) and BRCA2 (4% vs 1%; OR 3.09, 95% CI 1.49-6.55, q=0.1), than HER2-0 tumors. In HR-negative HER2-2+ tumors, a higher frequency of PIK3CA mutations was observed in comparison to HER2-0 (36.9% vs 19.5%; OR 2.41, 1.4-4.1, q=0.1), overall and in the metastatic setting. Conclusions: HER2-low BCs seem not to represent a distinct pathologic subtype. At the genomic level, however, some differences were identified and these became more conspicuous upon subclassification of HER2 IHC expression into 1+ and 2+. Further investigation into methods that more accurately detect and quantify low levels of HER2 expression in BC samples as well as better characterize the biology behind the HER2-low/ultralow expression is warranted.

Citation Format: Antonio Marra, Anton Safonov, Joshua Drago, Emanuela Ferraro, Pier Selenica, Andrea Gazzo, Giuseppe Curigliano, Shanu Modi, Pedram Razavi, Jorge Reis-Filho, Sarat Chandarlapaty. HER2-07 Genomic Characterization of Primary and Metastatic HER2-low Breast Cancers [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 HER2-07.

Abstract P4-02-01: Efficacy of HER2 ADCs against HER2 inhibitor resistance alterations in the PI3K and MAPK pathways in HER2-positive breast cancer

Background: HER2 positive (HER2+) breast cancers harboring downstream MAPK or PI3K pathway alterations manifest persistent downstream signaling on anti-HER2 inhibitors with metastatic patients having worse outcomes on first line trastuzumab and pertuzumab (HP) therapy. However, HER2 antibody-drug conjugates (ADCs) are not as dependent upon potent signal transduction inhibition to exert their antitumor effects. To further investigate, we sought to determine whether MAPK and/or PI3K alterations affect the biologic or clinical outcomes of patients and models receiving HER2 ADCs. Methods: We performed prospective genomic sequencing using MSK-IMPACT on patients with advanced HER2+ breast cancer who received trastuzumab emtansine (T-DM1) in the metastatic setting between March 2013 and July 2021. We collected detailed information on clinical outcomes and correlates through our institutional IRB-approved retrieval process. HER2/ER/PR status at the time of metastatic recurrence were defined as per ASCO/CAP guidelines. Cox proportional hazard models were used to determine the association between MAPK and PI3K pathways alterations and progression-free survival (PFS) on T-DM1. Common mutations associated with outcomes were modeled in HER2+ breast cancer cell lines using short hairpin RNAs and CRISPR/Cas9, and the sensitivity to HER2 ADC was evaluated via cell proliferation and xenograft assays. Results: We identified 185 HER2+ breast cancer patients treated with T-DM1 at any line (median: 5) whose primary (N=65) or metastatic (N=120) tumor samples were sequenced. Median age was 55 (range: 20-87). The majority of the patients received T-DM1 in 2nd or 3rd line (52%) and received prior trastuzumab or HER2 TKI in metastatic setting (96%). 74/185 (40%) had de novo metastatic breast cancer and 119/185 (64%) had ER/PR+/HER2+ disease. Pathogenic activating alterations involving the MAPK pathway were observed in 14% of patients with the most frequent alterations being ERBB2 activating mutations (42%) and NF1 loss (34%). PI3K pathway alterations were identified in 42% of the patients, the majority being activating mutations of PIK3CA (87%). MAPK alterations were significantly enriched in the metastatic tumors compared to the treatment-naïve primaries (20% vs 3%, p=0.001), while PI3K alterations were not (44% vs 40%, p=0.6). To reduce the possible confounding resistance mechanisms induced by prior treatment, we restricted the survival analyses to patients who received T-DM1 up to 3rd line of therapy (N=100). On multivariable analysis adjusted for ER/PR status (positive vs negative), stage at the presentation of metastatic disease (de novo metastatic vs recurrence), treatment line and type of sequenced sample (primary vs metastatic), patients with MAPK (N=14) and PI3K (N= 38) alterations had similar PFS compared to wild type (HR 1.20, 95%CI 0.62-2.30, p=0.6 and HR 1.23, 95%CI 0.77-1.95, p=0.4, respectively). Similar results were found in the combined analysis including alterations in either pathway (N=48, HR 1.28, 95%CI 0.81-2.04, p=0.3). To verify the antiproliferative effect of HER2 ADCs on breast cancer cells with MAPK pathway activation, we depleted NF1 in a panel of HER2+ breast cancer cell lines. Consistently, MAPK-altered cell lines were sensitive to FDA-approved HER2 ADCs including trastuzumab deruxtecan (T-DXd). Conclusions: In contrast to H/P therapy, T-DM1 therapy was equally effective in tumors with downstream PI3K or MAPK alterations and wild type tumors. Expanded analysis on a larger cohort, including a subgroup of patients treated with novel HER2 ADCs such as T-DXd will be presented. The characterization of PI3K and MAPK pathways status in metastatic HER2+ breast cancer may inform prioritization of treatment options.

Citation Format: Emanuela Ferraro, Anton Safonov, Yuan Chen, Charlie White, Antonio Marra, Mehnaj Ahmed, Barbara Acevedo, Chau T Dang, Shanu Modi, David B. Solit, Larry Norton, Mark E. Robson, Jorge Reis-Filho, Sarat Chandarlapaty, Pedram Razavi. Efficacy of HER2 ADCs against HER2 inhibitor resistance alterations in the PI3K and MAPK pathways in HER2-positive 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 P4-02-01.

Abstract P5-02-33: Deep learning-based assessment of HER2-low expression on breast cancer H&E digital whole slide images

Background: Antibody drug conjugates (ADCs) against HER2 have shown meaningful clinical activity in HER2 low breast cancers, defined as 1+ or 2+ staining on immunohistochemistry (IHC) without gene amplification by in situ hybridization (ISH) techniques. Given that these methods were originally developed for an accurate detection of HER2 3+, their sensitivity and robustness for the detection of low and ultra-low levels of HER2 are questionable. We have recently described a deep learning algorithm that can detect signatures of HER2 expression based on training utilizing scanned H&E whole slide images (WSI) of breast cancers for which IHC and mRNA expression levels of HER2 were available. Here, we report the application of our algorithm to two independent breast cancer cohorts. Methods: A model was developed based on recognition of invasive breast cancer in whole slide images of H&E staining, and then trained via computational neural network with multiple instance learning for binary classification of cases as HER2 “negative” and HER2 “expressed” (low). For training, true negatives were defined as having HER2 IHC-0 and mRNA level < 7.6. HER2-low cases were defined as IHC-1+/2+ and mRNA >9. IHC-0 cases with mRNA >7.6 were excluded from the training cohorts. The resulting model (HER2Complete) was able to distinguish HER2-negatives from HER2-low cases with an AUC of 0.91 (+/- 0.08). Here we use Her2Complete to assess HER2 in two additional cohorts that include 901 ER+/HER2 IHC-0 and 52 HER2 IHC 0+ breast cancers from MSK and TCGA cohorts, respectively. For the TCGA cohort, concomitant transcriptomics data (RNASeq) as a reference for HER2 mRNA expression were retrieved and “HER2 expressed” defined as RNASeq expression of HER2 greater than the 90th percentile of the geometric mean of expression of three reference genes not expressed in breast tissues (TTN, MUC13, OR10A6). Values less than this reference cut-off in the TCGA cohort were considered “HER2 not expressed.” Results: Among the 901 IHC-0 test cases from the MSK cohort, the model identified 82 as ‘negative’, whereas 819 were found to have features of HER2 expression (HER2-Low). Of the 82 negative cases in the MSK cohort, all except 13 cases expressed mRNA levels < 9, and 786/819 of the HER2-low cases expressed mRNA levels >8. Of the 52 IHC 0+ cases in the TCGA cohort, 33 also had “HER2 not expressed” by our reference based RNASeq expression cut-off. Our model identified 15 of these 33 as ‘negative’, while 15 of the 19 TCGA cases with IHC 0+ and HER2 ‘expressed’ by our cut-off were identified as ‘HER2-Low’ by our model. Conclusions: AI tools based on the analysis of WSIs of routinely prepared H&E sections may predict HER2 status in breast cancer. This work requires further investigation using treatment response data to demonstrate that cases with morphologic features of low level HER2 expression will respond to ADCs.

Citation Format: Gerard Oakley III, Jorge Reis-Filho, David Klimstra, Marc Goldfinger, Yikan Wang, Antonio Marra. Deep learning-based assessment of HER2-low expression on breast cancer H&E digital whole slide images [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 P5-02-33.

Abstract P1-13-17: Hyperactivation of the EGFR pathway is associated with resistance to tucatinib in HER2-positive breast cancer models

Background: The HER2-specific tyrosine kinase inhibitor (TKI) tucatinib (Tuca) recently approved for advanced HER2+ breast cancer is making a move towards the early setting. Given its growing use, resistance is inevitable, as observed in the HER2CLIMB study, where only one patient with brain metastasis remained progression free after 2 years on Tuca. Driven by the prevailing lack of knowledge about the mechanisms of resistance, in this study, we sought to define these mechanisms and identify treatment strategies to overcome them. We previously reported (SABCS 2021) that our BT474 TucaR models acquired EGFR amplification and showed elevated levels of phosphorylated (p) and total (t) EGFR, pHER2, pHER3, and downstream pAKT and pS6. Since the HER pathway is activated by ligands, here we aim to assess if hyperactivation of EGFR via high levels of its ligands is an alternative mechanism of Tuca resistance. Materials and Methods: Our recently developed HER2+ BT474 (ATCC and AZ) cell models with acquired resistance to Tuca (TucaR) developed through long-term exposure to gradually increasing doses of Tuca and their naïve parental (P) were used. Genomic (DNA-seq), transcriptomic (RNA-seq), and proteomic (western blot) characterization were performed. Changes in cell growth and migration were assessed by methylene blue and Incucyte wound healing assays, respectively. Results: RNA-seq analysis demonstrated that the levels of TGFα was significantly higher in our BT474 TucaR models compared to P cells. Our results now demonstrate that exogenous supplementation of EGF to BT474-P cells rescues the Tuca-mediated inhibition of pEGFR, pHER2, and the downstream pAKT, pERK, and pS6 levels. Exogenous EGF was also found to reduce the levels of apoptosis, as assessed by cleaved PARP, mitigating the Tuca-induced cell death. Exogenous EGF or TGFα rendered naïve BT474 and SKBR3 cells resistant to Tuca while neratinib, a pan-HER TKI, effectively inhibited this ligand-driven cell growth. We previously showed that the HER signaling reactivation observed in our EGFR-amplified TucaR cells was inhibited by the EGFR-specific TKI gefitinib (Gef) (SABCS 2021) and that the TucaR cells displayed enhanced migratory capabilities (AACR 2022). Here, we demonstrate that in addition to curbing the growth of TucaR cells, Gef, either alone or together with Tuca, also markedly reverts the migration of the TucaR cells. Knockdown (KD) of EGFR but not HER2 selectively and substantially inhibited the migration of the TucaR cells. KD of EGFR also had a marked cell killing effect on only the TucaR cells, whereas HER2 KD inhibited the growth of P but not TucaR cells. Our findings are consistent with the notion that while the P cells are functionally dependent on HER2, in TucaR cells the survival dependence could be rewired to rely primarily on the hyperactive EGFR signaling. Genomic analysis further revealed that in addition to EGFR amplification, the AZ TucaR cells also acquired a gain of YES1, a src family receptor tyrosine kinase implicated in cancer cell growth, invasion, and metastasis. Functional studies using 2 siRNAs, however, showed that YES1 KD had no effect on the growth of TucaR cells, and the migration of both TucaR and P cells was equally affected by YES1 KD, precluding the potential role of YES1 in driving the resistant and enhanced migratory phenotypes. Conclusions: Hyperactivation of the EGFR pathway via amplification of EGFR or increased expression of its ligands confers resistance to Tuca, which may be overcome using dual/pan-HER TKIs or the combination of potent EGFR and HER2 inhibitors. Given the rapidly evolving treatment landscape of HER2+ breast cancer and biomarkers of resistance, our novel findings have potentially crucial therapeutic implications and suggest that rationally sequencing the currently available TKIs may be clinically important.

Citation Format: Fu-Tien Liao, Tia Gordon, Chia Chia Liu, Pier Selenica, Yingjie Zhu, Juber Patel, Sarmistha Nanda, Lanfang Qin, Xiaoyong Fu, Andrea Gazzo, Antonio Marra, Juan Blanco-Heredia, Britta Weigelt, Jorge Reis-Filho, C. Kent Osborne, Mothaffar Rimawi, Rachel Schiff, Jamunarani Veeraraghavan. Hyperactivation of the EGFR pathway is associated with resistance to tucatinib in HER2-positive breast cancer models [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 P1-13-17.

Abstract P4-08-12: Clinical and Genomic Landscape of Breast Cancers Carrying CCNE1 Amplification

Background: Cyclin E1 (CCNE1) plays a critical role in cell cycle regulation. CCNE1 overexpression and/or gene amplification (amp) have been associated with poor outcome in several tumors, including breast cancer (BC). CCNE1 amp has recently been identified as a potential therapeutic target for novel synthetic lethality-based therapies. Here we sought to define the clinical and genomic features of BCs carrying CCNE1 amp. Methods: Genomic and clinical data from all consecutive BCs, which had been subjected to targeted sequencing using the FDA-authorized MSK-IMPACT assay from April 2014 to December 2021, were retrieved. Allele-specific copy number and fraction genome altered (FGA) were assessed using FACETS. Whole genome doubling (WGD) status was inferred from MSK-IMPACT sequencing data. Samples were categorized as CCNE1 amp or non-amp based on copy number profile assessed by FACETS. Mutual exclusivity and co-occurrence analyses between CCNE1 amp and other genetic alterations were performed using CoMEt. Multiple testing correction using the Benjamini–Hochberg procedure was applied to control for the false discovery rate. Progression-free survival (PFS) was assessed by Kaplan Meier method and Cox proportional-hazards models. Survival analyses were restricted to only patients with available pre-treatment samples. Results: Of 3,753 BCs with full clinical and genomic data, 125 (3.3%) harbored CCNE1 amp. A significant difference in the proportion of CCNE1 amp between treatment-naïve and post-treatment/metastatic samples was observed (2.4% vs 4%, p=0.007). CCNE1 amp was significantly less frequently detected in hormone receptor (HR)+/HER2- BCs than in HR-/HER2+ and HR-/HER2- subtypes (2% vs 7.6% and 7.2%, respectively, p< 0.001), and was particularly rare in invasive lobular BCs (1/452 cases). BCs with CCNE1 amp displayed a higher frequency of WGD (p< 0.001) and higher median FGA (p< 0.001) than non-amp tumors, overall and in different subtypes, suggesting increased genomic instability. No difference in tumor mutational burden (TMB) between CCNE1 amp and non-amp was found. In primary BC (n=1,385), a higher proportion of TP53 alterations was found in cases with CCNE1 amp (odds ratio [OR] 6.0, 95% confidence interval [CI] 2.5-16.6, q< 0.001). Conversely, CCNE1 amp was mutually exclusive with CDH1 alterations (q< 0.001). Comparable results were found in the analysis of post-treatment/metastatic samples (n=2,368). A subset analysis on HR+/HER2- BCs confirmed that TP53 (OR 4.2, 95%CI 2.28-8.11, q< 0.001) and CDH1 (OR 0.09, 95%CI 0.002-0.57, q< 0.1) alterations co-occurred and were mutually exclusive, respectively, with CCNE1 amp. ARID2 alterations were also enriched in HR+/HER2- tumors harboring CCNE1 amp (OR 10.6, 95%CI 2.54-33.93, q< 0.1). CCNE1 amp was significantly associated with reduced median PFS (8.8 vs 15.2 months in CCNE1 amp [n=9] vs CCNE1 non-amp [n=402]; hazard ratio [HR] 2.82, 95% CI 1.38-5.75, p=0.004) on first line treatment with CDK4/6 inhibitor plus endocrine therapy (ET) in HR+/HER2- metastatic BCs, regardless of the ET partner, FGA and TMB. CCNE1 amp was also associated with numerically inferior median PFS (7.3 vs 20.8 months in CCNE1 amp [n=5] vs CCNE1 non-amp [n=106]; HR 3.1, 95% CI 1.24-7.87, p=0.01) on first line trastuzumab/pertuzumab/taxane treatment in HER2+ metastatic BCs, with a trend toward significance after adjusting for FGA and TMB (p=0.09). Conclusions: CCNE1 amp is associated with specific clinicopathological and genomic features in BCs and linked to an increased genomic instability. CCNE1 amp defines a subset of metastatic BCs with marked poor clinical response to available standard-of-care treatments. Further studies testing novel therapeutic approaches, including synthetic lethality-based strategies targeting CCNE1 amp and CDK2-selective inhibition, are warranted.

Citation Format: Antonio Marra, Pier Selenica, Yingjie Zhu, Pedram Razavi, Anton Safonov, Emanuela Ferraro, Sarat Chandarlapaty, Jorge Reis-Filho. Clinical and Genomic Landscape of Breast Cancers Carrying CCNE1 Amplification [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 P4-08-12.

Abstract P4-02-18: Impact of clonal hematopoiesis on disease progression following CDK4/6 inhibitor therapy

Background Clonal Hematopoiesis (CH) is a well-established risk factor for adverse clinical outcomes including all-cause mortality, cardiovascular disease, and progression to hematologic malignancies. The presence of CH has been shown to adversely impact overall survival in non-hematologic cancers, however whether CH modulates response to specific therapies in breast cancer is not known. Here we investigate the impact of CH mutations on disease progression in patients with metastatic estrogen receptor (ER) positive breast cancer undergoing treatment with first line CDK4/6 inhibitors and endocrine therapy (CDK4/6i+ET). Methods We analyzed data from a well annotated cohort of patients with ER+ breast cancer who received endocrine therapy and CDK4/6 inhibitors. All patients underwent prospective tumor and matched WBC sequencing utilizing the MSK-IMPACT assay. CH variants were detected in blood samples utilizing the well-validated variant detection and filtration pipeline of MSK-IMPACT. CH mutations were defined as putative drivers (CH-PD) or non-putative drivers (CH) as previously described. To ensure the presence of CH at the time of therapy initiation, only patients who had CH sampling performed from 6 months before through 4 months after initiation of CDK4/6i+ET were included. We compared progression free survival (PFS) in patients with and without CH, as well as by CH-PD status and DNMT3A CH mutations. We investigated clinical covariates including type of endocrine therapy, receipt of prior neoadjuvant or adjuvant chemotherapy, and age at start of CDK4/6i+ET. Results The final cohort was comprised of 378 patients, of whom 135 (35.7%) had CH. The median time between sample collection and CDK4/6i+ET initiation was 0 (IQR -0.79 to 0.47 months). Patients with CH were older at time of therapy initiation (median 63.0 versus 54.7 years, p < 0.001). There were no significant differences between groups in terms of endocrine therapy (aromatase inhibitor or fulvestrant), prior chemotherapy, and time from CH sample collection to CDK4/6i+ET start. Univariate Cox-proportional hazard analysis did not reveal a difference between progression free survival and overall CH (HR 0.96, 95% CI 0.75 – 1.23, p = 0.76), CH-PD (HR 1.05, 0.77 – 1.43, p = 0.77), or DNMT3A mutations (HR 1.12, 0.80 – 1.60, p = 0.52) compared to patients without CH. Interestingly, age less than 60 years was found to be associated with PFS outcome (univariate HR 1.57, 1.22 – 2.01, p = 0.0004). Multivariate analysis adjusted for endocrine therapy partner and age at CDK4/6i+ET therapy did not reveal an association between outcome and overall CH (HR 1.07, 0.83 – 1.39, p = 0.59). In patients younger than age 60, presence of overall CH did not confer a significant PFS difference (HR 0.90, 0.63 – 1.29, p = 0.57). In the subset of patients older than 60 (n = 168) presence of CH conferred numerically, but not statistically, significant shorter PFS (HR 1.41 [0.96 – 2.09], p = 0.08). In this population, CH-PD conferred a shorter PFS (HR 1.75, 1.12 – 2.72, p = 0.02). Conclusion We found that CH, CH-PD and DNMT3A CH mutations did not affect PFS among metastatic ER+ breast cancer patients treated with first line CDK4/6 inhibitors. Younger age was associated with increased risk of progression, warranting further investigation. In the subset of patients with age older than 60, CH-PD conferred a shorter PFS. Further data, incorporating records of dose reductions, will be presented at the meeting.

Citation Format: Jacqueline Tao, Pablo Sanchez Vela, Anton Safonov, Emanuela Ferraro, Sebastia Franch Exposito, Kamal Menghrajani, Ryan Ptashkin, Elizabeth Comen, Lior Z. Braunstein, Mark E. Robson, Sarat Chandarlapaty, Jorge Reis-Filho, Michael Berger, Ahmet Zehir, Larry Norton, Ross Levine, Pedram Razavi. Impact of clonal hematopoiesis on disease progression following CDK4/6 inhibitor therapy [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 P4-02-18.

Abstract PD4-07: PD4-07 Uncovering molecular heterogeneity of mixed ductal and lobular carcinoma using digital spatial profiling

Background Mixed invasive ductal and lobular carcinoma (mDLC) is a rare special subtype (3-6%, ~10,000 cases/annually in US) of invasive breast cancer with elusive pathophysiology. This entity exhibits a mix of ductal-like and lobular-like tumor sub-components within the same tumor. With few seminal studies, mDLC remains poorly understood with little molecular understanding of its sub-components including their origin and implications on disease evolution, prognosis, and treatment response. With increasing recognition of no special type (NST) and invasive lobular carcinoma (ILC) as distinct diseases with unique biology, it is important to understand whether this mixed entity, and its sub-components are like NST and ILC subtypes or a distinct disease. Methods We identified mDLC cases from the UPMC cancer registry. These underwent comprehensive evaluation by a panel of expert pathologists. Three cases (each with a ductal and lobular sub-component on the same FFPE block) were shortlisted. These cases underwent digital spatial profiling (DSP) using Nanostring GeoMX Human Whole Transcriptome Atlas. Briefly, 5um slides were stained using RNAscope morphology marker probes (E-cadherin and PanCK) and GeoMX DSP oligo-conjugated RNA detection probes. Between 3-6 ductal and lobular regions of interest (ROI) per tumor were selected by pathologists. DSP barcodes were cleaved off using UV light and collected into 96-well plate. These underwent library preparation and sequencing. Raw reads were aligned to reference probes to quantify RNA counts. Q3 normalized counts were used in downstream analyses using R version 4.1. Linear modeling was used to assess differentially expressed genes (DEGs). Hypergeometric enrichment tests were used for geneset enrichment. T-tests was used to compare gene expression between two groups. Results In total 26 ROIs (14 ductal and 15 lobular) were profiled across the three mDLC FFPE slides. Overall data quality was excellent with > 90% sequencing saturation across profiled ROIs. Principle component analysis and consensus clustering showed that lobular and ductal ROIs clustered separately indicating distinct molecular profiles. Similarly, PAM50 analysis showed that ductal and lobular ROIs within each patient tumor had distinct PAM50 subtypes. To further investigate the molecular differences between ductal vs lobular ROIs, we performed differential gene expression analysis. We identified 38 up-regulated and 78 down-regulated genes in lobular compared to ductal ROIs. To assess whether mDLC sub-components share any molecular similarities to pure counterparts i.e., ILC and NST, we compared mDLC lobular vs ductal DEGs with those from TCGA ILC vs NST comparison. SHROOM1, KLK10 and KLK11 were up-regulated while CDH1, DCD, and CPB1 were down-regulated in both mDLC lobular ROIs and ILC vs mDLC ductal ROIs and NST, respectively. Pathway analysis revealed estrogen response, adhesion, and metabolism related differences between mDLC lobular vs ductal ROIs. Furthermore, key transcription factor signatures enriched in the up-regulated genes in lobular vs ductal ROIs included ESR1, FOXA2, GATA1/2 and AR signatures while those enriched in the down-regulated genes in lobular vs ductal ROIs included RCOR1, MYC, ZBTB7A, NELFE, and SPI1 signatures. Conclusion and Future Work Using DSP, we uncovered the molecular heterogeneity of mDLC. We revealed that lobular and ductal sub-components have distinct biology with differences in transcriptional signatures and hormone signaling, adhesion and metabolism related pathways. Our pilot study is the first to shed light on this elusive mixed entity using spatial profiling. Our future work will focus on DNA sequencing of mDLC sub-components to identify sub-component specific driver mutations. Our findings will need further investigation in larger mDLC cohorts to better understand their clinical implications in terms of evolution of this disease and its prognosis.

Citation Format: Osama Shiraz Shah, Azadeh Nasrazadani, Jennifer M. Atkinson, Celina Kleer, Priscilla F. McAuliffe, Rohit Bhargava, Jorge Reis-Filho, Peter C. Lucas, ADRIAN V. LEE, Steffi Oesterreich. PD4-07 Uncovering molecular heterogeneity of mixed ductal and lobular carcinoma using digital spatial profiling [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 PD4-07.

Abstract P2-09-03: Quantification of HER2 expression and spatial biology using computational pathology: A cross-assay validation study in breast cancer

Background Conventional pathologic scoring of HER2 by IHC is proven to distinguish potential responders to trastuzumab but has not been effective for next generation antibody drug conjugates (ADCs) such as trastuzumab deruxtecan (T-DXd), which is capable of bystander killing. Several alternative approaches have been deployed to measure HER2, including immunofluorescence and mRNA sequencing. We have developed a novel and fully automated computational pathology technique, Quantitative Continuous Scoring (QCS), to quantify the level and distribution of HER2 from digitized HER2 IHC slides in an objective, quantifiable, and reproducible manner on a per-cell basis [Gustavson et al., SABCS 2020]. To further validate this approach, we performed a systematic multi-omic comparison of QCS to orthogonal methods of HER2 quantitation on a cohort of primary and metastatic breast cancer cases (N=30). Methods HER2 was evaluated using three independent methods on serial tissue sections obtained from 30 archival FFPE breast cancer samples distributed over the full range of HER2 expression, from 0 to 3+. HER2-IHC staining (clone 4B5, Roche Tissue Diagnostics) was performed using standard methods and cases were scored by two pathologists using CAP/ASCO guidelines and H-scores were assigned. We performed FISH (HER2 IQFISH pharmDx [Dako]; PathVysion HER-2 DNA Probe Kit [Vysis]), mRNA quantification of ERBB2 transcript levels (Nano String), and immunofluorescence (IF; HER2 clone 29D8, CST). Imaged with Vectra (Akoya) and analyzed with Halo (Indica). QCS readouts were generated from the above-mentioned digital images of IHC slides by using a fully automated image analysis pipeline; readouts included per-cell staining intensity measurements of membranes and cytoplasmic sub-compartments in terms of optical density (OD) [Van der Laak, JQCS 2000], which were aggregated to a single slide-level score. Additionally, using the OD measurements and the cell locations, a Spatial Proximity Score (SPS) was computed, summing the percentage of cells with OD≥10 (corresponding to the limit of visual detection of IHC staining) as well as the percentage of cells with OD< 10 within a prespecified radius (25µm) of a neighboring cell with OD≥10. Results Our analysis demonstrated that QCS-based scoring correlates with orthogonal measurements used in this study. Comparing protein-based assays, the observed Pearson correlation was R=0.88 between QCS median membrane OD and IHC H-scores, R=0.86 with IF-based HER2 mean cell expression intensity, and R=0.85 with IF-based H-scores. Correlation with transcriptomic profiling was R=0.81 for OD vs. mRNA, however ERBB2 transcript levels did not distinguish between HER2 0, 1+, and 2+ FISH negative cases, while QCS was able to do so. Correlation between protein-based and nucleic-acid based assays were numerically worse, with R=0.64 for OD vs. FISH. All samples (including those with HER2 IHC scores of 0 and H-Scores < 10) had at least ~20% of cells with quantifiable HER2 expression by OD, the presence of which was confirmed using IF. For cases in the lowest quartile of HER2 expression by OD, SPS identified 20-50% additional HER2-null cells that were in close proximity to HER2-expressing cells that may be vulnerable to bystander killing. Conclusion QCS-based scoring is consistent with orthogonal protein-based measurements across the range of HER2 expression. Most importantly, QCS derived-spatial analysis features identify additional patients in the lower end of HER2 expression that might be highly relevant for ADC response prediction, particularly if a drug exerts bystander activity. Further clinical verification and validation on large cohorts is needed. Footnote: This study was approved by the IRB at MSKCC.

Citation Format: Joshua Drago, Zonera Hassan, Jan Zaucha, Ansh Kapil, Fatemeh Derakhshan, Fresia Pareja, Shimulov Anatoliy, Fanni Ratzon, Travis J Hollman, Claire Myers, Jessica Chan, Andrea Spitzmuller, Mark Gustavson, Danielle Carroll, Dara Ross, Jorge Reis-Filho, Carl barrett, Sihem Khalifa, Schmidt Guenter, Hadassah Sade, Sarat Chandarlapaty. Quantification of HER2 expression and spatial biology using computational pathology: A cross-assay validation study in 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 P2-09-03.

Abstract 2949: Enabling single cell analysis of copy number variation in breast cancer

Breast cancer is arguably one of the solid tumors most comprehensively analyzed by massively parallel sequencing. Although the repertoire of clonal somatic genetic alterations in primary and metastatic breast cancers has been characterized, a complete characterization of the subclonal genetic alterations present in these cancers, as well as their clonal composition and clonal dynamics remain to be fully defined. Addressing these knowledge gaps is germane to the successful personalization of treatments for breast cancer patients. Single cell sequencing is uncovering many insights into the clonal evolution and heterogeneity of breast cancers and how this can be leveraged for clinical and therapeutic benefit. Bulk next generation sequencing has primarily focused on mutation profiling; however, there are several lines of evidence to demonstrate the biological and clinical importance of accurately detecting copy number alterations. Previous single cell sequencing studies focusing on copy number alterations were based on whole-genome amplification methods or could not concurrently define the repertoire of somatic mutations and copy number alterations in single cells. Here we report on a novel single cell DNA sequencing approach leveraging the Tapestri device and Insights bioinformatic pipeline to detect both copy number alterations and somatic mutations in individual cells derived from breast cancer specimens. The identification of targeted somatic genetic alterations at single cell resolution provides novel insights about the development, progression and therapeutic resistance of breast cancers, and enables the development of novel diagnostic tools and predictive biomarkers.

Citation Format: Saurabh Parikh, Arnaud Da Cruz Paula, Shirin Issa Bhaloo, Alex Li, Britta Weigelt, Khushali Patel, Ania Wronski, Anup Parikh, Khushali Patel, Jorge Reis-Filho. Enabling single cell analysis of copy number variation in breast cancer [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 2949.

Abstract P2-13-19: Statin modulation of antibody drug conjugate activity in breast cancer models and patients

Background: Membrane HER2 expression levels strongly influence the activity of anti-HER2 therapies directed at the HER2 extracellular domain (1,2). Tumoral cell-surface caveolin-1 (CAV-1) regulates receptor tyrosine kinase membrane trafficking mechanisms (1,3). In xenograft models, CAV-1 can be modulated with cholesterol-depleting drugs, such as statins. We hypothesized that preclinical and clinical use of statins might influence the expression of CAV-1 and thereby affect the efficacy of anti-HER2 antibody-drug conjugates (ADCs) in breast cancer. Methods: Preclinically, statins were given alone and in combination with HER2 antibody drug conjugates such as T-DM1 to assess HER2 levels, drug uptake, and antitumor efficacy in HER2-positive cancer models. Mice received an intravenous injection of T-DM1, oral doses of lovastatin, or a combination of T-DM1 and lovastatin, for 5 weeks. Clinically, we performed retrospective analyses of HER2-positive MBC patients at MSKCC who received T-DM1 and consented to molecular profiling and clinical data abstraction. Progression-free survival (PFS) on T-DM1 was estimated using Kaplan Meyer methods and compared using the log rank test. Univariable and multivariable Cox proportional hazards models were constructed using relevant clinical covariates. Results: Tumor models with high levels of CAV-1 exhibit low HER2 density on the cell surface and show low T-DM1 targeted immunoPET binding when compared with CAV-1-low tumors. Mechanistic studies showed that CAV-1 depletion in HER2-positive BC cells temporally stabilizes HER2 on the surface and delays T-DM1 recycling. In preclinical BC models, CAV-1 depletion induced by synthetic oligonucleotides or statins enhances T-DM1 binding and efficacy in tumors with incomplete HER2 membranous reactivity. For clinical analyses, a total of 164 patients who received T-DM1 were included in the final analysis. Twenty-one (12.8%) of these patients were recorded to be taking statins concurrently with T-DM1, as part of their routine clinical care. Compared with patients with no recorded statin use, patients receiving statins were older on average (median age 58 vs. 50 years, p = 0.007), but did not differ in their race, histologic breast cancer type, ER status, or T-DM1 treatment line. Median PFS on T-DM1 in the overall cohort was 5.5 months. Median PFS in patients who received statins was 14 months vs. 5.4 months in patients who had no recorded statin use (p = 0.1). In univariable Cox analysis, the association between statin use and PFS did not reach statistical significance (p=0.119). In a multivariable analysis including age, ER status, treatment line, and breast cancer histology, no variable reached statistical significance. However, statin use had the greatest observed degree of association with PFS (p = 0.17). Conclusions: Statins modulate surface HER2 levels and T-DM1 efficacy preclinically via CAV-1. Although only a numerical difference in outcomes was observed in the MSKCC cohort, the effect appeared meaningful and therefore assessment in larger patient cohorts is now underway. Based on the results of these forthcoming analyses, prospective trials may be justified that integrate these well-tolerated and low-cost agents into HER2 ADC treatment regimens. References:1-Pereira P. et al Nat Commun 9, 5137 (2018)2-Chew H.Y. et al Cell 180, 895 (2020)3-Pereira P. et al Clin Cancer Res 26, 6215 (2020)

Citation Format: Joshua Z Drago, Patricia R Pereira, Anton Safonov, Antonio Marra, Yi Rao, Bo Liu, Mehnaj Ahmed, Shanu Modi, Jorge Reis-Filho, Mark Robson, Filippo Montemurro, Pedram Razavi, Jason S Lewis, Sarat Chandarlapaty. Statin modulation of antibody drug conjugate activity in breast cancer models and patients [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 P2-13-19.

HER2 + breast cancers evade anti-HER2 therapy via a switch in driver pathway

Inhibition of HER2 in HER2-amplified breast cancer has been remarkably successful clinically, as demonstrated by the efficacy of HER-kinase inhibitors and HER2-antibody treatments. Whilst resistance to HER2 inhibition is common in the metastatic setting, the specific programs downstream of HER2 driving resistance are not established. Through genomic profiling of 733 HER2-amplified breast cancers, we identify enrichment of somatic alterations that promote MEK/ERK signaling in metastatic tumors with shortened progression-free survival on anti-HER2 therapy. These mutations, including NF1 loss and ERBB2 activating mutations, are sufficient to mediate resistance to FDA-approved HER2 kinase inhibitors including tucatinib and neratinib. Moreover, resistant tumors lose AKT dependence while undergoing a dramatic sensitization to MEK/ERK inhibition. Mechanistically, this driver pathway switch is a result of MEK-dependent activation of CDK2 kinase. These results establish genetic activation of MAPK as a recurrent mechanism of anti-HER2 therapy resistance that may be effectively combated with MEK/ERK inhibitors.

Abstract 2488: Characterization of gene fusions in paired primary and metastatic samples of breast cancer in the AURORA molecular screening program

Introduction

Gene fusions (GFs) are genomic alterations with oncogenic potential. This class of alterations is highly actionable with targeted therapies showing high rates of durable responses in agnostic settings such as tumors with NTRK fusions. Recurrent GFs are rare in metastatic breast cancer (mBC) and ESR1 fusions drive endocrine resistance. AURORA is a molecular screening program for patients with mBC early in the course of their disease, collecting primary tumors, metastatic samples, liquid samples and longitudinal clinical follow-up data.

Experimental procedures

Patients are enrolled at the diagnosis of metastatic disease or after 1 line of therapy. Targeted gene sequencing of 411 BC genes is performed on the primary tumor, a metastatic sample, whole blood for germline variants and, for a subset of these genes, on tumor DNA from a baseline plasma sample. RNA-seq is performed on primary and metastatic tumor tissue samples. GFs were identified by rnafusion, a pipeline implementing 5 widely-used tools. Only GFs detected by at least 3 tools or 2 tools and 1 GF database match were considered for downstream analysis. Selected GFs were validated in-silico by FusionInspector. GFs detected in independent human normal tissue datasets were filtered out. Findings were correlated to available genomics, transcriptomics and clinical data.

Results

GFs data were generated from 316 paired primary/metastatic samples from 158 patients with curated clinical and genomic data: 97 ER+/HER2- (61.4%), 37 triple-negative (23.4%) and 24 HER2+ (15.2%). A total of 538 fusions were called in the primary samples (mean = 3.4 per sample) and 707 in the metastatic samples (mean = 4.5) with a validation rate of 72%. The gene fusion burden in metastatic samples was higher than in the primary (p<1e-4), especially for ER+ patients. In metastatic samples, gene fusion burden by PAM50 subtype was higher for basal, HER2-E and LumB when compared to LumA (max p=2e-3). A total of 400 acquired GFs were identified (mean=2.5 per sample), involving relevant BC genes, such as ESR1, ERBB2, NF1 and FGFR1.The presence of fusions involving actionable genes was associated with a shorter progression-free survival (PFS) on therapies in the metastatic setting (median PFS=8 vs 17 months, p=0.001).

Conclusions

We report on the characterization of GFs in a large cohort of patients with mBC. Through the analysis of matched primary and metastatic tumor samples from 158 patients, we delineated the landscape of acquired gene fusions in BC. We observed a significant increase of gene fusion burden in metastatic compared to corresponding primary samples, involving key BC genes. Fusions involving actionable genes were associated with shorter PFS. Additional integrative analyses combining detected GFs and available genomics, gene expression and/or patients' treatments and outcomes are ongoing and will be presented during the meeting.

Citation Format: Matteo Benelli, Chiara Biagioni, Danai Fimereli, Florentine S. Hilbers, Claudia De Angelis, Ana Vivancos, David Venet, Andrea Vingiani, Alexandre Irrthum, Veerle Van Dooren, Peter Willem Vuylsteke, Sonia Servitja, Jorge Reis-Filho, Giuseppe Curigliano, Mafalda Oliveira, Martine Piccart, Philippe Aftimos. Characterization of gene fusions in paired primary and metastatic samples of breast cancer in the AURORA molecular screening program [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 2488.

Pitfalls in assessing stromal tumor infiltrating lymphocytes (sTILs) in breast cancer

Stromal tumor-infiltrating lymphocytes (sTILs) are important prognostic and predictive biomarkers in triple-negative (TNBC) and HER2-positive breast cancer. Incorporating sTILs into clinical practice necessitates reproducible assessment. Previously developed standardized scoring guidelines have been widely embraced by the clinical and research communities. We evaluated sources of variability in sTIL assessment by pathologists in three previous sTIL ring studies. We identify common challenges and evaluate impact of discrepancies on outcome estimates in early TNBC using a newly-developed prognostic tool. Discordant sTIL assessment is driven by heterogeneity in lymphocyte distribution. Additional factors include: technical slide-related issues; scoring outside the tumor boundary; tumors with minimal assessable stroma; including lymphocytes associated with other structures; and including other inflammatory cells. Small variations in sTIL assessment modestly alter risk estimation in early TNBC but have the potential to affect treatment selection if cutpoints are employed. Scoring and averaging multiple areas, as well as use of reference images, improve consistency of sTIL evaluation. Moreover, to assist in avoiding the pitfalls identified in this analysis, we developed an educational resource available at www.tilsinbreastcancer.org/pitfalls.

Pan-cancer analysis of whole genomes

Cancer is driven by genetic change, and the advent of massively parallel sequencing has enabled systematic documentation of this variation at the whole-genome scale1-3. Here we report the integrative analysis of 2,658 whole-cancer genomes and their matching normal tissues across 38 tumour types from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA). We describe the generation of the PCAWG resource, facilitated by international data sharing using compute clouds. On average, cancer genomes contained 4-5 driver mutations when combining coding and non-coding genomic elements; however, in around 5% of cases no drivers were identified, suggesting that cancer driver discovery is not yet complete. Chromothripsis, in which many clustered structural variants arise in a single catastrophic event, is frequently an early event in tumour evolution; in acral melanoma, for example, these events precede most somatic point mutations and affect several cancer-associated genes simultaneously. Cancers with abnormal telomere maintenance often originate from tissues with low replicative activity and show several mechanisms of preventing telomere attrition to critical levels. Common and rare germline variants affect patterns of somatic mutation, including point mutations, structural variants and somatic retrotransposition. A collection of papers from the PCAWG Consortium describes non-coding mutations that drive cancer beyond those in the TERT promoter4; identifies new signatures of mutational processes that cause base substitutions, small insertions and deletions and structural variation5,6; analyses timings and patterns of tumour evolution7; describes the diverse transcriptional consequences of somatic mutation on splicing, expression levels, fusion genes and promoter activity8,9; and evaluates a range of more-specialized features of cancer genomes8,10-18.

Abstract A21: Modeling of oncogenic chromosomal translocations of aggressive fusion-positive sarcomas by CRISPR-Cas9 genomic engineering

A prominent challenge in fusion-positive sarcoma research is that the oncogenic driver is typically a chimeric transcription factor. Although considered the disease-specific molecular drivers, these chimeric transcription factors remain challenging to target pharmacologically. An alternative approach is the elucidation of molecular pathways controlled by the oncogenic transcription factor, with the aim of identifying druggable targets.

However, many fusion-driven sarcomas have a paucity of suitable genetic models, with some lacking any patient-derived cell lines. In the situations where patient-derived cell lines are available, most have been passaged for decades, potentially adding further complexity and bias. Therefore, development of faithful model systems of oncogenic chromosomal translocations is of principal importance for these aggressive translocation-driven pediatric sarcomas.

Here, we describe a novel approach that combines CRISPR-Cas9 genomic editing technology, with homology-directed repair (HDR) to engineer, capture, and modulate the expression of chromosomal translocation products in a human cell line.

We have applied this approach to the genetic modeling of t(11;22)(q24;q12) and t(11;22)(p13;q12), translocation products of the EWSR1 gene and its 3 fusion partners FLI1 and WT1, present in Ewing's sarcoma and desmoplastic small round cell tumor, respectively. Our approach establishes an innovative platform for constructing isogenic and conditionally inducible biologically relevant models for a variety of sarcomas driven by chromosomal translocations.

Citation Format: Lee Spraggon, Luciano Martelotto, Julija Hmeljak, Tyler Hitchman, Jiang Wang, Lu Wang, Emily Slotkin, Pang-Dian Fan, Jorge Reis-Filho, Marc Ladanyi. Modeling of oncogenic chromosomal translocations of aggressive fusion-positive sarcomas by CRISPR-Cas9 genomic engineering [abstract]. In: Proceedings of the AACR Conference on Advances in Sarcomas: From Basic Science to Clinical Translation; May 16-19, 2017; Philadelphia, PA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(2_Suppl):Abstract nr A21.

Assessing Tumor-Infiltrating Lymphocytes in Solid Tumors: A Practical Review for Pathologists and Proposal for a Standardized Method from the International Immuno-Oncology Biomarkers Working Group: Part 2: TILs in Melanoma, Gastrointestinal Tract Carcinomas, Non-Small Cell Lung Carcinoma and Mesothelioma, Endometrial and Ovarian Carcinomas, Squamous Cell Carcinoma of the Head and Neck, Genitourinary Carcinomas, and Primary Brain Tumors

Assessment of the immune response to tumors is growing in importance as the prognostic implications of this response are increasingly recognized, and as immunotherapies are evaluated and implemented in different tumor types. However, many different approaches can be used to assess and describe the immune response, which limits efforts at implementation as a routine clinical biomarker. In part 1 of this review, we have proposed a standardized methodology to assess tumor-infiltrating lymphocytes (TILs) in solid tumors, based on the International Immuno-Oncology Biomarkers Working Group guidelines for invasive breast carcinoma. In part 2 of this review, we discuss the available evidence for the prognostic and predictive value of TILs in common solid tumors, including carcinomas of the lung, gastrointestinal tract, genitourinary system, gynecologic system, and head and neck, as well as primary brain tumors, mesothelioma and melanoma. The particularities and different emphases in TIL assessment in different tumor types are discussed. The standardized methodology we propose can be adapted to different tumor types and may be used as a standard against which other approaches can be compared. Standardization of TIL assessment will help clinicians, researchers and pathologists to conclusively evaluate the utility of this simple biomarker in the current era of immunotherapy.

Patient-derived xenograft (PDX) models in basic and translational breast cancer research

Patient-derived xenograft (PDX) models of a growing spectrum of cancers are rapidly supplanting long-established traditional cell lines as preferred models for conducting basic and translational preclinical research. In breast cancer, to complement the now curated collection of approximately 45 long-established human breast cancer cell lines, a newly formed consortium of academic laboratories, currently from Europe, Australia, and North America, herein summarizes data on over 500 stably transplantable PDX models representing all three clinical subtypes of breast cancer (ER+, HER2+, and "Triple-negative" (TNBC)). Many of these models are well-characterized with respect to genomic, transcriptomic, and proteomic features, metastatic behavior, and treatment response to a variety of standard-of-care and experimental therapeutics. These stably transplantable PDX lines are generally available for dissemination to laboratories conducting translational research, and contact information for each collection is provided. This review summarizes current experiences related to PDX generation across participating groups, efforts to develop data standards for annotation and dissemination of patient clinical information that does not compromise patient privacy, efforts to develop complementary data standards for annotation of PDX characteristics and biology, and progress toward "credentialing" of PDX models as surrogates to represent individual patients for use in preclinical and co-clinical translational research. In addition, this review highlights important unresolved questions, as well as current limitations, that have hampered more efficient generation of PDX lines and more rapid adoption of PDX use in translational breast cancer research.

Assessing Tumor-infiltrating Lymphocytes in Solid Tumors: A Practical Review for Pathologists and Proposal for a Standardized Method From the International Immunooncology Biomarkers Working Group: Part 1: Assessing the Host Immune Response, TILs in Invasive Breast Carcinoma and Ductal Carcinoma In Situ, Metastatic Tumor Deposits and Areas for Further Research

Assessment of tumor-infiltrating lymphocytes (TILs) in histopathologic specimens can provide important prognostic information in diverse solid tumor types, and may also be of value in predicting response to treatments. However, implementation as a routine clinical biomarker has not yet been achieved. As successful use of immune checkpoint inhibitors and other forms of immunotherapy become a clinical reality, the need for widely applicable, accessible, and reliable immunooncology biomarkers is clear. In part 1 of this review we briefly discuss the host immune response to tumors and different approaches to TIL assessment. We propose a standardized methodology to assess TILs in solid tumors on hematoxylin and eosin sections, in both primary and metastatic settings, based on the International Immuno-Oncology Biomarker Working Group guidelines for TIL assessment in invasive breast carcinoma. A review of the literature regarding the value of TIL assessment in different solid tumor types follows in part 2. The method we propose is reproducible, affordable, easily applied, and has demonstrated prognostic and predictive significance in invasive breast carcinoma. This standardized methodology may be used as a reference against which other methods are compared, and should be evaluated for clinical validity and utility. Standardization of TIL assessment will help to improve consistency and reproducibility in this field, enrich both the quality and quantity of comparable evidence, and help to thoroughly evaluate the utility of TILs assessment in this era of immunotherapy.

Abstract 2607: The cBioPortal for Cancer Genomics: an open source platform for accessing and interpreting complex cancer genomics data in the era of precision medicine

The cBioPortal for Cancer Genomics is an open-access portal (http://cbioportal.org) that enables interactive, exploratory analysis of large-scale cancer genomics data. It integrates genomic and clinical data, and provides a suite of visualization and analysis options, including cohort and patient-level visualization, mutation visualization, survival analysis, enrichment analysis, and network analysis. The user interface is user-friendly, responsive, and makes genomic data easily accessible to translational scientists, biologists, and clinicians.

The cBioPortal is a fully open source platform. All code is available on GitHub (https://github.com/cBioPortal/) under GNU Affero GPL license. The code base is maintained by multiple groups, including Memorial Sloan Kettering Cancer Center, Dana-Farber Cancer Institute, Children’s Hospital of Philadelphia, Princess Margaret Cancer Centre, and The Hyve, an open source bioinformatics company based in the Netherlands. More than 30 academic centers as well as multiple pharmaceutical and biotech companies maintain private instances of the cBioPortal. This includes the recently launched cBioPortal instance at the NCI Genomic Data Commons (https://cbioportal.gdc.nci.nih.gov/), and two large cBioPortal instances hosting genomic and clinical data at MSK and DFCI, supporting the MSK-IMPACT and DFCI Profile projects, two of the largest clinical sequencing efforts in the world.

Our multi-institutional software team has accelerated the progress of evolving the core architectural technologies and developing new features to keep pace with the rapidly advancing fields of cancer genomics and precision cancer medicine. For example, we have integrated multi-platform genomics data with extensive clinical data including patient demographics, treatment history, and survival data. We have also developed a patient-centric view that visualizes both clinical and genomic data with annotation from OncoKB knowledge base. In the next few years, the development team will focus on the following areas:

(1) Implementing major architectural changes to ensure future scalability and performance.

(2) New features to support precision medicine, including (i) improved integration of knowledge base annotation, (ii) enhanced visualization of patient timeline, drug response, and tumor evolution, (iii) new patient similarity metrics, (iv) improved support for immunogenomics and immunotherapy, and (v) new visualization and analysis features for understanding response to therapy.

(3) New analysis and target discovery features for large cohorts, including (i) supporting user-defined virtual cohort by selecting samples from multiple studies, and (ii) comparison of genomic or clinical characteristics of two or more selected cohorts.

(4) Expanding community outreach, user support and training, and documentation.

Citation Format: Jianjiong Gao, Ersin Ciftci, Pichai Raman, Pieter Lukasse, Istemi Bahceci, Adam Abeshouse, Hsiao-Wei Chen, Ino de Bruijn, Benjamin Gross, Zachary Heins, Ritika Kundra, Aaron Lisman, Angelica Ochoa, Robert Sheridan, Onur Sumer, Yichao Sun, Jiaojiao Wang, Manda Wilson, Hongxin Zhang, James Xu, Andy Dufilie, Priti Kumari, James Lindsay, Anthony Cros, Karthik Kalletla, Fedde Schaeffer, Sander Tan, Sjoerd van Hagen, Jorge Reis-Filho, Kees van Bochove, Ugur Dogrusoz, Trevor Pugh, Adam Resnick, Chris Sander, Ethan Cerami, Nikolaus Schultz. The cBioPortal for Cancer Genomics: an open source platform for accessing and interpreting complex cancer genomics data in the era of precision medicine [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2607. doi:10.1158/1538-7445.AM2017-2607

PALB2, CHEK2 and ATM rare variants and cancer risk: data from COGS

BACKGROUND

The rarity of mutations in PALB2, CHEK2 and ATM make it difficult to estimate precisely associated cancer risks. Population-based family studies have provided evidence that at least some of these mutations are associated with breast cancer risk as high as those associated with rare BRCA2 mutations. We aimed to estimate the relative risks associated with specific rare variants in PALB2, CHEK2 and ATM via a multicentre case-control study.

METHODS

We genotyped 10 rare mutations using the custom iCOGS array: PALB2 c.1592delT, c.2816T>G and c.3113G>A, CHEK2 c.349A>G, c.538C>T, c.715G>A, c.1036C>T, c.1312G>T, and c.1343T>G and ATM c.7271T>G. We assessed associations with breast cancer risk (42 671 cases and 42 164 controls), as well as prostate (22 301 cases and 22 320 controls) and ovarian (14 542 cases and 23 491 controls) cancer risk, for each variant.

RESULTS

For European women, strong evidence of association with breast cancer risk was observed for PALB2 c.1592delT OR 3.44 (95% CI 1.39 to 8.52, p=7.1×10-5), PALB2 c.3113G>A OR 4.21 (95% CI 1.84 to 9.60, p=6.9×10-8) and ATM c.7271T>G OR 11.0 (95% CI 1.42 to 85.7, p=0.0012). We also found evidence of association with breast cancer risk for three variants in CHEK2, c.349A>G OR 2.26 (95% CI 1.29 to 3.95), c.1036C>T OR 5.06 (95% CI 1.09 to 23.5) and c.538C>T OR 1.33 (95% CI 1.05 to 1.67) (p≤0.017). Evidence for prostate cancer risk was observed for CHEK2 c.1343T>G OR 3.03 (95% CI 1.53 to 6.03, p=0.0006) for African men and CHEK2 c.1312G>T OR 2.21 (95% CI 1.06 to 4.63, p=0.030) for European men. No evidence of association with ovarian cancer was found for any of these variants.

CONCLUSIONS

This report adds to accumulating evidence that at least some variants in these genes are associated with an increased risk of breast cancer that is clinically important.

Abstract 5277: The cBioPortal for cancer genomics and its application in precision oncology

The cBioPortal for Cancer Genomics provides intuitive visualization and analysis of complex cancer genomics data. The public site (http://cbioportal.org/) is accessed by more than 1,500 researchers per day, and there are now dozens of local instances of the software that host private data sets at cancer centers around the globe.

We have recently released the software under an open source license, making it free to use and modify by anybody. The software and detailed documentation are available at https://github.com/cBioPortal/cbioportal.

We are now establishing a multi-institutional software development network, which will coordinate and drive the future development of the software and associated data pipelines. This group will focus on four main areas:

1. New analysis and visualization features, including:

a. Improved support for cross-cancer queries and cohort comparisons.

b. Enhanced clinical decision support for precision oncology, including an improved patient view with knowledge base integration, patient timelines and improved tools for visualizing tumor evolution.

2. New data pipelines, including support for new genomic data types and streamlined pipelines for TCGA and the International Cancer Genome Consortium (ICGC).

3. Software architecture and performance improvements.

4. Community engagement: Documentation, user support, and training.

This coordinated effort will help to further establish the cBioPortal as the software of choice in cancer genomics research, both in academia and the pharmaceutical industry. Furthermore, as the sequencing of tumor samples has entered clinical practice, we are expanding the features of the software so that it can be used for precision medicine at cancer centers. In particular, clean, web-accessible, interactive clinical reports integrating multiple sources of genome variation and clinical annotation over time has potential to improve clinical action beyond current text-based molecular reports. By making complex genomic data easily interpretable and linking it to information about drugs and clinical trials, the cBioPortal software has the potential to facilitate the use of genomic data in clinical decision making.

Citation Format: Jianjiong Gao, James Lindsay, Stuart Watt, Istemi Bahceci, Pieter Lukasse, Adam Abeshouse, Hsiao-Wei Chen, Ino de Bruijn, Benjamin Gross, Dong Li, Ritika Kundra, Zachary Heins, Jorge Reis-Filho, Onur Sumer, Yichao Sun, Jiaojiao Wang, Qingguo Wang, Hongxin Zhang, Priti Kumari, M. Furkan Sahin, Sander de Ridder, Fedde Schaeffer, Kees van Bochove, Ugur Dogrusoz, Trevor Pugh, Chris Sander, Ethan Cerami, Nikolaus Schultz. The cBioPortal for cancer genomics and its application in precision oncology. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5277.

Abstract 1470: Ovarian cancer PDXs preserve preexisting genetic oligoclonality

Advanced Epithelial Ovarian Cancer (EOC) is the leading cause of female cancer casualties and its prognosis remains grim, with about 30% of 5-year overall survival. Classically EOCs have been stratified according to 5 histotypes, tumor grading and disease stages. However, worst prognosis is associated to high grade EOCs, which mostly comprise serous ovarian carcinomas (SOC). A large majority of the patients relapse by 18 to 24 months after end of chemotherapy (CT) and eventually develop resistant disease. Research aiming at proposing more effective therapies relies on a limited set of cancer cell lines established several decades ago and whose representativeness of the actual disease has been put in doubt (Domcke et al., 2013; Nature Comm).

This strongly calls for the production of novel EOC models to support experimental and preclinical work. In the last decade patient derived xenografts (PDX) have gained considerable interest, because they faithfully reproduce the characteristics of the tumor of origin.

We, here, report the establishment of a collection of 35 ovarian cancer PDXs resulting from the original graft of 77 ovarian tumor samples onto immuno-compromised mice. Interestingly, established PDXs covered the diversity of EOC histotypes. We characterized a subset of 14 EOC PDXs at the genetic (array-CGH and transcriptome) and histological levels. This revealed that PDXs perfectly reproduce the original genetic and morphological features of the ovarian tumors they stemmed from. It was of note that PDXs conserved the principal characteristics of the original CNC profiles over several passages. However, a detailed comparison revealed fluctuation of CNC patterns in a fraction of chromosomal regions between the original tumor and the PDXs, that could be attributed to the oligoclonal nature of the original tumor. The analysis by CGH, FISH and exome sequencing of one case, for which several tumor nodules were sampled and grafted, revealed that PDX globally maintained an oligoclonal structure. Despite fluctuations in the relative prevalence of some mutations according to the sample, we did not observe any overgrowth of a particular subclone present in the original tumor. This suggests that xenotransplantation of ovarian tumors and growth as PDX is not associated to a drastic change in selective pressure. We believe these findings are of interest to determine the impact of different therapy regimen on the clonality of EOC.

Citation Format: Pierre-Emmanuel Colombo, Stanislas du Manoir, Beatrice Orsetti, Rui Bras-Goncalves, Alan MacKay, Maryou Lambros, Tuan T. Nguyen, Florence Boissiere, Didier Pourquier, Jorge Reis-Filho, Charles G. Theillet. Ovarian cancer PDXs preserve preexisting genetic oligoclonality. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1470. doi:10.1158/1538-7445.AM2015-1470

Abstract PD1-4: Somatic leukemogenic mutations associated with infiltrating white blood cells in breast cancer patients

Background: In the last few decades, theoretical models of cancer growth and progression have long focused on the aberrations of cancer cells alone, such as the abnormal mitotic and invasive characteristics of cancer cells. More recent research across multiple solid tumors suggests a critical interplay between solid tumors and immune regulating cells. Mounting evidence suggests that the immune system can tip the scales of cancer progression, eliciting either an anti-tumor or pro-tumor immune response depending upon varying stimulating and inhibitory factors. Here, we are the first to demonstrate novel mutations including leukemogenic mutations among tumor infiltrating lymphocytes in breast cancer patients.

Methods: We obtained 17 primary breast cancer samples from patients who presented for either a lumpectomy or mastectomy as part of an IRB approved biospecimen protocol. Of the 17 patient samples, 13 had triple negative breast cancer, 2 had ER+, HER2+ disease, and 2 had ER+, HER2- disease. In the 17 samples, we used fluorescent activated cell sorting to separate CD45-positive hematopoietic cells from CD45-negative epithelial cells. We then performed exome sequencing of tumor-infiltrating hematopoietic cells to investigate for the presence of pathogenic mutations in tumor-associated leukocytes. In this first step, we identified candidate mutations in known cancer genes, including BCOR, NOTCH2, TET2, NF1, EZH2, and JAK1. As a validation step, we then performed capture-based sequencing of tumor-infiltrating leukocytes in 20 breast cancer samples matched to each patient’s germline DNA sample (buccal swab). In 10 of the 20 patients, we identified and validated somatic mutations. Of note, 6 of these patients harbored mutations known to be associated with leukemia, including DNTM3A, TET2, and BCOR. Most of these mutations were present in at least 5-20% of reads. This suggests that these mutations were present in enriched subclones and were not rare alleles occurring in a minority of hematopoietic stem cells. Lastly, we performed 454 deep sequencing analysis of microdissected tumor DNA samples and confirmed the absence of these mutations in breast cancer cells.

Conclusion: Our data demonstrate somatic mutations in tumor infiltrating leukocytes in breast tumors which were not identified in matched germline or tumor DNA samples. Notably, some of these mutations have been implicated in the pathogenesis of lymphoid and myeloid malignancies. This observation suggests a unique relationship between cancer cells and mutant infiltrating leukocytes. We are now investigating the functional interaction between cancer cells and hematopoietic cells. Our findings reframe our understanding of carcinogenesis and offer novel opportunities for cancer detection and treatment.

Citation Format: Elizabeth A Comen, Maria Kleppe, Hannah Wen, Britta Weigelt, Lennart Bastian, Brian Blum, Franck T Rapaport, Matt Keller, Nicolas Socci, Agnes Viale, Daoqi You, Robert Benezra, Edi Brogi, Jorge Reis-Filho, Michael Berger, Ross Levine, Larry Norton. Somatic leukemogenic mutations associated with infiltrating white blood cells in breast cancer patients [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr PD1-4.

Molecular characterisation of isogenic taxane resistant cell lines identify novel drivers of drug resistance

Background

Taxanes such as paclitaxel and docetaxel are used successfully to treat breast cancer, usually in combination with other agents. They interfere with microtubules causing cell cycle arrest; however, the mechanisms underlying the clinical effects of taxanes are yet to be fully elucidated.

Methods

Isogenic paclitaxel resistant (PACR) MDA‒MB‒231, paclitaxel resistant ZR75‒1 and docetaxel resistant (DOCR) ZR75‒1 cell lines were generated by incrementally increasing taxane dose in native cell lines in vitro. We used aCGH analysis to identify mechanisms driving taxane resistance.

Results

Taxane resistant cell lines exhibited an 18-170 fold increased resistance to taxanes, with the ZR75-1 resistant cell lines also demonstrating cross resistance to anthracyclines. Paclitaxel treatment of native cells resulted in a G2/M block and a decrease in the G1 phase of the cell cycle. However, in the resistant cell lines, minimal changes were present. Functional network analysis revealed that the mitotic prometaphase was lost in the resistant cell lines.

Conclusion

This study established a model system for examining taxane resistance and demonstrated that both MDR and mitosis represent common mechanism of taxane resistance.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2407-14-762) contains supplementary material, which is available to authorized users.

Inference of tumor evolution during chemotherapy by computational modeling and in situ analysis of genetic and phenotypic cellular diversity

Cancer therapy exerts a strong selection pressure that shapes tumor evolution, yet our knowledge of how tumors change during treatment is limited. Here, we report the analysis of cellular heterogeneity for genetic and phenotypic features and their spatial distribution in breast tumors pre- and post-neoadjuvant chemotherapy. We found that intratumor genetic diversity was tumor-subtype specific, and it did not change during treatment in tumors with partial or no response. However, lower pretreatment genetic diversity was significantly associated with pathologic complete response. In contrast, phenotypic diversity was different between pre- and posttreatment samples. We also observed significant changes in the spatial distribution of cells with distinct genetic and phenotypic features. We used these experimental data to develop a stochastic computational model to infer tumor growth patterns and evolutionary dynamics. Our results highlight the importance of integrated analysis of genotypes and phenotypes of single cells in intact tissues to predict tumor evolution.

Abstract P4-05-10: PIK3CA mutations are enriched in invasive lobular carcinomas and invasive mammary carcinomas with lobular features: Results from a TCGA sub-analysis

Background: Aberrant signaling via the PI3K pathway is a common alteration in breast cancer (BC), with frequent activating mutations in the PIK3CA gene helical (exon 9) and catalytic (exon 20) domains. These mutations occur across all BC subtypes with an overall incidence of 36%, with the highest frequency (∼45%) in luminal A/ER+ tumors. Lobular phenotype is common among luminal A tumors. We examined associations between lobular histology and molecular features among BC samples submitted for comprehensive molecular analyses for The Cancer Genome Atlas (TCGA).

Design: Experts in breast pathology reviewed digital slides of breast cancer samples submitted for comprehensive molecular profiling to the TCGA. Tumors were graded, subtyped and scored for additional histopathologic features. We tested pairwise associations between lobular features and components of grade, PAM50-derived molecular subtype and mutational status for BRAC1/2, PIK3CA, TP53 and CDH1 by performing Chi-Square analysis for comparisons with a categorical variable and the Mann-Whitney test for comparisons with an ordinal variable

Results: A total of 1132 images were scored from 589 unique cases in TCGA. For cases with multiple scorers (43% of cases), we summarized scores by taking the median (for ordinal variables) or the consensus diagnosis (for categorical variables). A total of 567 cases had a consensus diagnosis for lobular features, all of which had pathological information on components of histologic grade and 540 of which had data for TP53, CDH1, and PIK3CA mutations. 110/567 (19%) of cases were classified as invasive lobular or invasive mammary carcinoma with lobular features. The lobular cases had significantly less nuclear pleomorphism (p = 3.3 e -12), lower mitotic index (p = 3.4e-16), less tubule formation (p = 3.9e-8), increased association with lobular carcinoma in situ (p < 2.2 e-16), decreased stromal inflammation (p = 1.5e-7), and decreased necrosis (p = 4.4e-11) compared with cases without lobular features. Cases with lobular features were highly enriched for CDH1 mutations with 19% of cases with lobular features having CDH1 mutations, compared with only 1% of cases without lobular features (p = 2.4 e-14). The lobular features cases were more likely to have PIK3CA mutations (p = 0.01), with 33% of the lobular features cases having PIK3CA mutations, compared with 21% of the non-lobular cases. The lobular features cases were less likely to have TP53 mutations (p = 0.02), with 13% of lobular features cases having TP53 mutations as compared with 24% of the non-lobular feature cases. Lobular status was associated with PAM50 molecular subtype (Chi-square p = 0.002) with the lobular cases significantly less likely to be basal molecular subtype and more likely to be Luminal-A.

Conclusions: PIK3CA mutations are enriched in invasive lobular carcinomas and invasive mammary carcinomas with lobular features. These associations point to the possibility that PIK3CA mutations as well as CDH1 alterations are important drivers of invasive lobular carcinomas.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P4-05-10.