Accelerating Drug Development Using Spatial Multi-omics

SUMMARY

Spatial biology approaches enabled by innovations in imaging biomarker platforms and artificial intelligence-enabled data integration and analysis provide an assessment of patient and disease heterogeneity at ever-increasing resolution. The utility of spatial biology data in accelerating drug programs, however, requires balancing exploratory discovery investigations against scalable and clinically applicable spatial biomarker analysis.

Post-therapy emergence of an NBN reversion mutation in a patient with pancreatic acinar cell carcinoma

Pancreatic acinar cell carcinoma (PACC) is a rare form of pancreatic cancer that commonly harbors targetable alterations, including activating fusions in the MAPK pathway and loss-of-function (LOF) alterations in DNA damage response/homologous recombination DNA repair-related genes. Here, we describe a patient with PACC harboring both somatic biallelic LOF of NBN and an activating NTRK1 fusion. Upon disease progression following 13 months of treatment with folinic acid, fluorouracil, irinotecan, and oxaliplatin (FOLFIRINOX), genomic analysis of a metastatic liver biopsy revealed the emergence of a novel reversion mutation restoring the reading frame of NBN. To our knowledge, genomic reversion of NBN has not been previously reported as a resistance mechanism in any tumor type. The patient was treated with, but did not respond to, targeted treatment with a selective NTRK inhibitor. This case highlights the complex but highly actionable genomic landscape of PACC and underlines the value of genomic profiling of rare tumor types such as PACC.

Hypoxia-Directed Treatment of Human Papillomavirus-Related Oropharyngeal Carcinoma

PURPOSE

Standard curative-intent chemoradiotherapy for human papillomavirus (HPV)-related oropharyngeal carcinoma results in significant toxicity. Since hypoxic tumors are radioresistant, we posited that the aerobic state of a tumor could identify patients eligible for de-escalation of chemoradiotherapy while maintaining treatment efficacy.

METHODS

We enrolled patients with HPV-related oropharyngeal carcinoma to receive de-escalated definitive chemoradiotherapy in a phase II study (ClinicalTrials.gov identifier: NCT03323463). Patients first underwent surgical removal of disease at their primary site, but not of gross disease in the neck. A baseline 18F-fluoromisonidazole positron emission tomography scan was used to measure tumor hypoxia and was repeated 1-2 weeks intratreatment. Patients with nonhypoxic tumors received 30 Gy (3 weeks) with chemotherapy, whereas those with hypoxic tumors received standard chemoradiotherapy to 70 Gy (7 weeks). The primary objective was achieving a 2-year locoregional control (LRC) of 95% with a 7% noninferiority margin.

RESULTS

One hundred fifty-eight patients with T0-2/N1-N2c were enrolled, of which 152 patients were eligible for analyses. Of these, 128 patients met criteria for 30 Gy and 24 patients received 70 Gy. The 2-year LRC was 94.7% (95% CI, 89.8 to 97.7), meeting our primary objective. With a median follow-up time of 38.3 (range, 22.1-58.4) months, the 2-year progression-free survival (PFS) and overall survival (OS) rates were 94% and 100%, respectively, for the 30-Gy cohort. The 70-Gy cohort had similar 2-year PFS and OS rates at 96% and 96%, respectively. Acute grade 3-4 adverse events were more common in 70 Gy versus 30 Gy (58.3% v 32%; P = .02). Late grade 3-4 adverse events only occurred in the 70-Gy cohort, in which 4.5% complained of late dysphagia.

CONCLUSION

Tumor hypoxia is a promising approach to direct dosing of curative-intent chemoradiotherapy for HPV-related carcinomas with preserved efficacy and substantially reduced toxicity that requires further investigation.

Image-based multiplex immune profiling of cancer tissues: translational implications. A report of the International Immuno-oncology Biomarker Working Group on Breast Cancer

Recent advances in the field of immuno-oncology have brought transformative changes in the management of cancer patients. The immune profile of tumours has been found to have key value in predicting disease prognosis and treatment response in various cancers. Multiplex immunohistochemistry and immunofluorescence have emerged as potent tools for the simultaneous detection of multiple protein biomarkers in a single tissue section, thereby expanding opportunities for molecular and immune profiling while preserving tissue samples. By establishing the phenotype of individual tumour cells when distributed within a mixed cell population, the identification of clinically relevant biomarkers with high-throughput multiplex immunophenotyping of tumour samples has great potential to guide appropriate treatment choices. Moreover, the emergence of novel multi-marker imaging approaches can now provide unprecedented insights into the tumour microenvironment, including the potential interplay between various cell types. However, there are significant challenges to widespread integration of these technologies in daily research and clinical practice. This review addresses the challenges and potential solutions within a structured framework of action from a regulatory and clinical trial perspective. New developments within the field of immunophenotyping using multiplexed tissue imaging platforms and associated digital pathology are also described, with a specific focus on translational implications across different subtypes of cancer. © 2024 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Converging and evolving immuno-genomic routes toward immune escape in breast cancer

The interactions between tumor and immune cells along the course of breast cancer progression remain largely unknown. Here, we extensively characterize multiple sequential and parallel multiregion tumor and blood specimens of an index patient and a cohort of metastatic triple-negative breast cancers. We demonstrate that a continuous increase in tumor genomic heterogeneity and distinct molecular clocks correlated with resistance to treatment, eventually allowing tumors to escape from immune control. TCR repertoire loses diversity over time, leading to convergent evolution as breast cancer progresses. Although mixed populations of effector memory and cytotoxic single T cells coexist in the peripheral blood, defects in the antigen presentation machinery coupled with subdued T cell recruitment into metastases are observed, indicating a potent immune avoidance microenvironment not compatible with an effective antitumor response in lethal metastatic disease. Our results demonstrate that the immune responses against cancer are not static, but rather follow dynamic processes that match cancer genomic progression, illustrating the complex nature of tumor and immune cell interactions.

Adenoid cystic carcinoma of the Bartholin's gland is underpinned by MYB- and MYBL1- rearrangements

OBJECTIVE

Adenoid cystic carcinoma (AdCC) of the Bartholin's gland (AdCC-BG) is a very rare gynecologic vulvar malignancy. AdCC-BGs are slow-growing but locally aggressive and are associated with high recurrence rates. Here we sought to characterize the molecular underpinning of AdCC-BGs.

METHODS

AdCC-BGs (n = 6) were subjected to a combination of RNA-sequencing, targeted DNA-sequencing, reverse-transcription PCR, fluorescence in situ hybridization (FISH) and MYB immunohistochemistry (IHC). Clinicopathologic variables, somatic mutations, copy number alterations and chimeric transcripts were assessed.

RESULTS

All six AdCC-BGs were biphasic, composed of ductal and myoepithelial cells. Akin to salivary gland and breast AdCCs, three AdCC-BGs had the MYB::NFIB fusion gene with varying breakpoints, all of which were associated with MYB overexpression by IHC. Two AdCC-BGs were underpinned by MYBL1 fusion genes with different gene partners, including MYBL1::RAD51B and MYBL1::EWSR1 gene fusions, and showed MYB protein expression. Although the final AdCC-BG studied had MYB protein overexpression, no gene fusion was identified. AdCC-BGs harbored few additional somatic genetic alterations, and only few mutations in cancer-related genes were identified, including GNAQ, GNAS, KDM6A, AKT1 and BCL2, none of which were recurrent. Two AdCC-BGs, both with a MYB::NFIB fusion gene, developed metastatic disease.

CONCLUSIONS

AdCC-BGs constitute a convergent phenotype, whereby activation of MYB or MYBL1 can be driven by the MYB::NFIB fusion gene or MYBL1 rearrangements. Our observations further support the notion that AdCCs, irrespective of organ site, constitute a genotypic-phenotypic correlation. Assessment of MYB or MYBL1 rearrangements may be used as an ancillary marker for the diagnosis of AdCC-BGs.

Genomic and epigenomic basis of breast invasive lobular carcinomas lacking CDH1 genetic alterations

CDH1 (E-cadherin) bi-allelic inactivation is the hallmark alteration of breast invasive lobular carcinoma (ILC), resulting in its discohesive phenotype. A subset of ILCs, however, lack CDH1 genetic/epigenetic inactivation, and their genetic underpinning is unknown. Through clinical targeted sequencing data reanalysis of 364 primary ILCs, we identified 25 ILCs lacking CDH1 bi-allelic genetic alterations. CDH1 promoter methylation was frequent (63%) in these cases. Targeted sequencing reanalysis revealed 3 ILCs harboring AXIN2 deleterious fusions (n = 2) or loss-of-function mutation (n = 1). Whole-genome sequencing of 3 cases lacking bi-allelic CDH1 genetic/epigenetic inactivation confirmed the AXIN2 mutation and no other cell-cell adhesion genetic alterations but revealed a new CTNND1 (p120) deleterious fusion. AXIN2 knock-out in MCF7 cells resulted in lobular-like features, including increased cellular migration and resistance to anoikis. Taken together, ILCs lacking CDH1 genetic/epigenetic alterations are driven by inactivating alterations in other cell adhesion genes (CTNND1 or AXIN2), endorsing a convergent phenotype in ILC.

Regression-based Deep-Learning predicts molecular biomarkers from pathology slides

Deep Learning (DL) can predict biomarkers from cancer histopathology. Several clinically approved applications use this technology. Most approaches, however, predict categorical labels, whereas biomarkers are often continuous measurements. We hypothesize that regression-based DL outperforms classification-based DL. Therefore, we develop and evaluate a self-supervised attention-based weakly supervised regression method that predicts continuous biomarkers directly from 11,671 images of patients across nine cancer types. We test our method for multiple clinically and biologically relevant biomarkers: homologous recombination deficiency score, a clinically used pan-cancer biomarker, as well as markers of key biological processes in the tumor microenvironment. Using regression significantly enhances the accuracy of biomarker prediction, while also improving the predictions' correspondence to regions of known clinical relevance over classification. In a large cohort of colorectal cancer patients, regression-based prediction scores provide a higher prognostic value than classification-based scores. Our open-source regression approach offers a promising alternative for continuous biomarker analysis in computational pathology.

Nonlobular Invasive Breast Carcinomas with Biallelic Pathogenic CDH1 Somatic Alterations: A Histologic, Immunophenotypic, and Genomic Characterization

CDH1 encodes for E-cadherin, and its loss of function is the hallmark of invasive lobular carcinoma (ILC). Albeit vanishingly rare, biallelic CDH1 alterations may be found in nonlobular breast carcinomas (NL-BCs). We sought to determine the clinicopathologic characteristics and repertoire of genetic alterations of NL-BCs harboring CDH1 biallelic genetic alterations. Analysis of 5842 breast cancers (BCs) subjected to clinical tumor-normal sequencing with an FDA-cleared multigene panel was conducted to identify BCs with biallelic CDH1 pathogenic/likely pathogenic somatic mutations lacking lobular features. The genomic profiles of NL-BCs with CDH1 biallelic genetic alterations were compared with those of ILCs and invasive ductal carcinomas (IDCs), matched by clinicopathologic characteristics. Of the 896 CDH1-altered BCs, 889 samples were excluded based on the diagnosis of invasive mixed ductal/lobular carcinoma or ILC or the detection of monoallelic CDH1 alterations. Only 7 of the 5842 (0.11%) BCs harbored biallelic CDH1 alterations and lacked lobular features. Of these, 4/7 (57%) cases were ER-positive/HER2-negative, 1/7 (14%) was ER-positive/HER2-positive, and 2/7 (29%) were ER-negative/HER2-negative. In total, 5/7 (71%) were of Nottingham grade 2, and 2/7 (29%) were of grade 3. The NL-BCs with CDH1 biallelic genetic alterations included a mucinous carcinoma (n = 1), IDCs with focal nested growth (n = 2), IDC with solid papillary (n = 1) or apocrine (n = 2) features, and an IDC of no special type (NST; n = 1). E-cadherin expression, as detected by immunohistochemistry, was absent (3/5) or aberrant (discontinuous membranous/cytoplasmic/granular; 2/5). However, NL-BCs with CDH1 biallelic genetic alterations displayed recurrent genetic alterations, including TP53, PIK3CA (57%, 4/7; each), FGFR1, and NCOR1 (28%, 2/7, each) alterations. Compared with CDH1 wild-type IDC-NSTs, NL-BCs less frequently harbored GATA3 mutations (0% vs 47%, P = .03), but no significant differences were detected when compared with matched ILCs. Therefore, NL-BCs with CDH1 biallelic genetic alterations are vanishingly rare, predominantly comprise IDCs with special histologic features, and have genomic features akin to luminal B ER-positive BCs.

Decreased HER2 expression in endometrial cancer following anti-HER2 therapy

Trastuzumab has demonstrated clinical efficacy in the treatment of HER2-positive serous endometrial cancer (EC), which led to its incorporation into standard-of-care management of this aggressive disease. Acquired resistance remains an important challenge, however, and its underlying mechanisms in EC are unknown. To define the molecular changes that occur in response to anti-HER2 therapy in EC, targeted next-generation sequencing (NGS), HER2 immunohistochemistry (IHC), and fluorescence in situ hybridization (FISH) were performed on pre- and post-treatment tumour samples from 14 patients with EC treated with trastuzumab or trastuzumab emtansine. Recurrent tumours after anti-HER2 therapy acquired additional genetic alterations compared with matched pre-treatment ECs and frequently showed decreased HER2 protein expression by IHC (7/14, 50%). Complete/near-complete absence of HER2 protein expression (score 0/1+) observed post-treatment (4/14, 29%) was associated with retained HER2 gene amplification (n = 3) or copy number neutral status (n = 1). Whole-exome sequencing performed on primary and recurrent tumours from the latter case, which exhibited genetic heterogeneity of HER2 amplification in the primary tumour, revealed selection of an early HER2-non-amplified clone following therapy. Our findings demonstrate that loss of target expression, by selection of HER2-non-amplified clones or, more commonly, by downregulation of expression, may constitute a mechanism of resistance to anti-HER2 therapy in HER2-positive EC. © 2023 The Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Standard-of-care systemic therapy with or without stereotactic body radiotherapy in patients with oligoprogressive breast cancer or non-small-cell lung cancer (Consolidative Use of Radiotherapy to Block [CURB] oligoprogression): an open-label, randomised, controlled, phase 2 study

BACKGROUND

Most patients with metastatic cancer eventually develop resistance to systemic therapy, with some having limited disease progression (ie, oligoprogression). We aimed to assess whether stereotactic body radiotherapy (SBRT) targeting oligoprogressive sites could improve patient outcomes.

METHODS

We did a phase 2, open-label, randomised controlled trial of SBRT in patients with oligoprogressive metastatic breast cancer or non-small-cell lung cancer (NSCLC) after having received at least first-line systemic therapy, with oligoprogression defined as five or less progressive lesions on PET-CT or CT. Patients aged 18 years or older were enrolled from a tertiary cancer centre in New York, NY, USA, and six affiliated regional centres in the states of New York and New Jersey, with a 1:1 randomisation between standard of care (standard-of-care group) and SBRT plus standard of care (SBRT group). Randomisation was done with a computer-based algorithm with stratification by number of progressive sites of metastasis, receptor or driver genetic alteration status, primary site, and type of systemic therapy previously received. Patients and investigators were not masked to treatment allocation. The primary endpoint was progression-free survival, measured up to 12 months. We did a prespecified subgroup analysis of the primary endpoint by disease site. All analyses were done in the intention-to-treat population. The study is registered with ClinicalTrials.gov, NCT03808662, and is complete.

FINDINGS

From Jan 1, 2019, to July 31, 2021, 106 patients were randomly assigned to standard of care (n=51; 23 patients with breast cancer and 28 patients with NSCLC) or SBRT plus standard of care (n=55; 24 patients with breast cancer and 31 patients with NSCLC). 16 (34%) of 47 patients with breast cancer had triple-negative disease, and 51 (86%) of 59 patients with NSCLC had no actionable driver mutation. The study was closed to accrual before reaching the targeted sample size, after the primary efficacy endpoint was met during a preplanned interim analysis. The median follow-up was 11·6 months for patients in the standard-of-care group and 12·1 months for patients in the SBRT group. The median progression-free survival was 3·2 months (95% CI 2·0-4·5) for patients in the standard-of-care group versus 7·2 months (4·5-10·0) for patients in the SBRT group (hazard ratio [HR] 0·53, 95% CI 0·35-0·81; p=0·0035). The median progression-free survival was higher for patients with NSCLC in the SBRT group than for those with NSCLC in the standard-of-care group (10·0 months [7·2-not reached] vs 2·2 months [95% CI 2·0-4·5]; HR 0·41, 95% CI 0·22-0·75; p=0·0039), but no difference was found for patients with breast cancer (4·4 months [2·5-8·7] vs 4·2 months [1·8-5·5]; 0·78, 0·43-1·43; p=0·43). Grade 2 or worse adverse events occurred in 21 (41%) patients in the standard-of-care group and 34 (62%) patients in the SBRT group. Nine (16%) patients in the SBRT group had grade 2 or worse toxicities related to SBRT, including gastrointestinal reflux disease, pain exacerbation, radiation pneumonitis, brachial plexopathy, and low blood counts.

INTERPRETATION

The trial showed that progression-free survival was increased in the SBRT plus standard-of-care group compared with standard of care only. Oligoprogression in patients with metastatic NSCLC could be effectively treated with SBRT plus standard of care, leading to more than a four-times increase in progression-free survival compared with standard of care only. By contrast, no benefit was observed in patients with oligoprogressive breast cancer. Further studies to validate these findings and understand the differential benefits are warranted.

FUNDING

National Cancer Institute.

Somatic estrogen receptor α mutations that induce dimerization promote receptor activity and breast cancer proliferation

Physiologic activation of estrogen receptor α (ERα) is mediated by estradiol (E2) binding in the ligand-binding pocket of the receptor, repositioning helix 12 (H12) to facilitate binding of coactivator proteins in the unoccupied coactivator binding groove. In breast cancer, activation of ERα is often observed through point mutations that lead to the same H12 repositioning in the absence of E2. Through expanded genetic sequencing of breast cancer patients, we identified a collection of mutations located far from H12 but nonetheless capable of promoting E2-independent transcription and breast cancer cell growth. Using machine learning and computational structure analyses, this set of mutants was inferred to act distinctly from the H12-repositioning mutants and instead was associated with conformational changes across the ERα dimer interface. Through both in vitro and in-cell assays of full-length ERα protein and isolated ligand-binding domain, we found that these mutants promoted ERα dimerization, stability, and nuclear localization. Point mutations that selectively disrupted dimerization abrogated E2-independent transcriptional activity of these dimer-promoting mutants. The results reveal a distinct mechanism for activation of ERα function through enforced receptor dimerization and suggest dimer disruption as a potential therapeutic strategy to treat ER-dependent cancers.

The role of APOBEC3B in lung tumor evolution and targeted cancer therapy resistance

In this study, the impact of the apolipoprotein B mRNA-editing catalytic subunit-like (APOBEC) enzyme APOBEC3B (A3B) on epidermal growth factor receptor (EGFR)-driven lung cancer was assessed. A3B expression in EGFR mutant (EGFRmut) non-small-cell lung cancer (NSCLC) mouse models constrained tumorigenesis, while A3B expression in tumors treated with EGFR-targeted cancer therapy was associated with treatment resistance. Analyses of human NSCLC models treated with EGFR-targeted therapy showed upregulation of A3B and revealed therapy-induced activation of nuclear factor kappa B (NF-κB) as an inducer of A3B expression. Significantly reduced viability was observed with A3B deficiency, and A3B was required for the enrichment of APOBEC mutation signatures, in targeted therapy-treated human NSCLC preclinical models. Upregulation of A3B was confirmed in patients with NSCLC treated with EGFR-targeted therapy. This study uncovers the multifaceted roles of A3B in NSCLC and identifies A3B as a potential target for more durable responses to targeted cancer therapy.

KIT genetic alterations in breast cancer

AIMS

Activating somatic mutations or gene amplification of KIT result in constitutive activation of its receptor tyrosine kinase, which is targetable in various solid tumours. Here, we sought to investigate the presence of KIT genetic alterations in breast cancer (BC) and characterise the histological and genomic features of these tumours.

METHODS

A retrospective analysis of 5,575 BCs previously subjected to targeted sequencing using the FDA-authorised Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Targets (MSK-IMPACT) assay was performed to identify BCs with KIT alterations. A histological assessment of KIT-altered BCs was conducted, and their repertoire of genetic alterations was compared with that of BCs lacking KIT genetic alterations, matched for age, histological type, oestrogen receptor/HER2 status and sample type.

RESULTS

We identified 18 BCs (0.32%), including 9 primary and 9 metastatic BCs, with oncogenic/likely oncogenic genetic alterations affecting KIT, including activating somatic mutations (n=4) or gene amplification (n=14). All KIT-altered BCs were of high histological grade, although no distinctive histological features were observed. When compared with BCs lacking KIT genetic alterations, no distinctive genetic features were identified. In two metastatic KIT-altered BCs in which the matched primary BC had also been analysed by MSK-IMPACT, the KIT mutations were found to be restricted to the metastatic samples, suggesting that they were late events in the evolution of these cancers.

CONCLUSIONS

KIT genetic alterations are vanishingly rare in BC. KIT-altered BCs are of high grade but lack distinctive histological features. Genetic alterations in KIT might be late events in the evolution and/or progression of BC.

Metaplastic Breast Cancer: Current Understanding and Future Directions

Metaplastic breast cancers (MBC) encompass a group of highly heterogeneous tumors which share the ability to differentiate into squamous, mesenchymal or neuroectodermal components. While often termed rare breast tumors, given the relatively high prevalence of breast cancer, they are seen with some frequency. Depending upon the definition applied, MBC represents 0.2% to 1% of breast cancers diagnosed in the United States. Less is known about the epidemiology of MBC globally, though a growing number of reports are providing information on this. These tumors are often more advanced at presentation relative to breast cancer broadly. While more indolent subtypes exist, the majority of MBC subtypes are associated with inferior survival. MBC is most commonly of triple-negative phenotype. In less common hormone receptor positive MBCs, hormone receptor status appears not to be prognostic. In contrast, relatively rare HER2-positive MBCs are associated with superior outcomes. Multiple potentially targetable molecular features are overrepresented in MBC including DNA repair deficiency signatures and PIK3/AKT/mTOR and WNT pathways alterations. Data on the prevalence of targets for novel antibody-drug conjugates is also emerging. While chemotherapy appears to be less active in MBC than in other breast cancer subtypes, efficacy is seen in some MBCs. Disease-specific trials, as well as reports of exceptional responses, may provide clues for novel approaches to this often hard-to-treat breast cancer. Strategies which harness newer research tools, such as large data and artificial intelligence hold the promise of overcoming historic barriers to the study of uncommon tumors and could markedly advance disease-specific understanding in MBC.

Histologic and genomic characterization of a primary mucinous carcinoma of the skin

Aims

Primary skin mucinous carcinoma is a rare sweat gland neoplasm with a high local recurrence rate after conventional excision but a low distant-metastasis rate. The genetic underpinning of skin mucinous carcinoma is presently unknown. Here, we sought to define whether the repertoire of somatic mutations of a primary mucinous carcinoma of the skin would be similar to that of mucinous breast carcinomas, given the histologic similarities between these tumor types.

Methods and results

The tumor was situated in the dermis and partially involved the subcutaneous fat. Tumor cells were suspended in periodic acid-Schiff diastaseresistant- positive mucin lakes and expressed cytokeratin 7, synaptophysin and estrogen receptor. DNA samples extracted from microdissected tumor and matched normal tissue were subjected to massively parallel sequencing targeting 410 cancer-related genes. The skin mucinous tumor was found to have a low tumor mutation burden, but to harbor a clonal GATA3 frameshift mutation (p. T418Hfs*89) and amplification of FOXA1, genes not uncommonly altered in breast mucinous carcinomas.

Conclusions

In this primary skin mucinous carcinoma, GATA3 and FOXA1 driver genetic events were identified, consistent with a possible developmental relationship between skin and breast mucinous neoplasms.

Genomic Profiling Reveals Germline Predisposition and Homologous Recombination Deficiency in Pancreatic Acinar Cell Carcinoma

PURPOSE

To determine the genetic predisposition underlying pancreatic acinar cell carcinoma (PACC) and characterize its genomic features.

METHODS

Both somatic and germline analyses were performed using an Food and Drug Administration-authorized matched tumor/normal sequencing assay on a clinical cohort of 28,780 patients with cancer, 49 of whom were diagnosed with PACC. For a subset of PACCs, whole-genome sequencing (WGS; n = 12) and RNA sequencing (n = 6) were performed.

RESULTS

Eighteen of 49 (36.7%) PACCs harbored germline pathogenic variants in homologous recombination (HR) and DNA damage response (DDR) genes, including BRCA1 (n = 1), BRCA2 (n = 12), PALB2 (n = 2), ATM (n = 2), and CHEK2 (n = 1). Thirty-one PACCs displayed pure, and 18 PACCs harbored mixed acinar cell histology. Fifteen of 31 (48%) pure PACCs harbored a germline pathogenic variant affecting HR-/DDR-related genes. BRCA2 germline pathogenic variants (11 of 31, 35%) were significantly more frequent in pure PACCs than in pancreatic adenocarcinoma (86 of 2,739, 3.1%; P < .001), high-grade serous ovarian carcinoma (67 of 1,318, 5.1%; P < .001), prostate cancer (116 of 3,401, 3.4%; P < .001), and breast cancer (79 of 3,196, 2.5%; P < .001). Genomic features of HR deficiency (HRD) were detected in 7 of 12 PACCs undergoing WGS, including 100% (n = 6) of PACCs with germline HR-related pathogenic mutations and 1 of 6 PACCs lacking known pathogenic alterations in HR-related genes. Exploratory analyses revealed that in PACCs, the repertoire of somatic driver genetic alterations and the load of neoantigens with high binding affinity varied according to the presence of germline pathogenic alterations affecting HR-/DDR-related genes and/or HRD.

CONCLUSION

In a large pan-cancer cohort, PACC was identified as the cancer type with the highest prevalence of both BRCA2 germline pathogenic variants and genomic features of HRD, suggesting that PACC should be considered as part of the spectrum of BRCA-related malignancies.

HER2 Genetic Intratumor Heterogeneity Is Associated With Resistance to Trastuzumab and Trastuzumab Emtansine Therapy in Recurrent High-Grade Endometrial Cancer

Anti-HER2 targeted therapies have recently demonstrated clinical activity in the treatment of high-grade endometrial carcinomas (ECs), particularly serous carcinomas with HER2 amplification and/or overexpression. Intratumor heterogeneity of HER2 amplification or HER2 genetic intratumor heterogeneity (G-ITH) has been associated with resistance to anti-HER2 therapies in breast and gastroesophageal cancers; however, its clinical relevance in EC is unknown. To characterize HER2 G-ITH in EC, archival specimens from a clinically annotated cohort of 57 ECs treated with trastuzumab or trasutuzmab emtansine in the recurrent (n = 38) or adjuvant (n = 19) setting were subjected to central pathology review, HER2 assessment by immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH), and next-generation sequencing. HER2 G-ITH, defined as HER2 amplification in 5% to 50% of tumor cells examined by FISH, was identified in 36% (19/53) of ECs and was associated with lower HER2 copy number and levels of protein expression. HER2 IHC revealed spatially distinct areas of strong expression juxtaposed with areas of low/absent expression in tumors with the "cluster" pattern of G-ITH, whereas the "mosaic" pattern was typically associated with a diffuse admixture of cells with variable levels of HER2 expression. HER2 G-ITH was frequently observed in cases with IHC/FISH or FISH/next-generation sequencing discrepancies and/or with an equivocal/negative FISH result (9/13, 69%). Although the objective response rate to anti-HER2 therapy in recurrent ECs was 52% (13/25) for tumors lacking HER2 G-ITH, none (0%, 0/10) of the patients with HER2 G-ITH achieved a complete or partial response (P = .005). HER2 G-ITH was significantly associated with worse progression-free survival (hazard ratio, 2.88; 95% CI, 1.33-6.27; P = .005) but not overall survival. HER2 IHC score, HER2/CEP17 ratio, HER2 copy number, histologic subtype, and other genetic alterations, including PIK3CA hotspot mutations, were not significantly associated with therapeutic response or survival outcomes. Treatment responses were not restricted to serous carcinomas, supporting consideration of anti-HER2 therapy in patients with HER2-positive high-grade ECs of non-serous histology. Our results demonstrate that HER2 G-ITH is an important determinant of response to trastuzumab and trastuzumab emtansine in EC, providing a rationale for the development of novel therapeutic strategies to target HER2-nonamplified resistant tumor subpopulations, such as HER2 antibody-drug conjugates with bystander effects.

Chronic Interferon Stimulated Gene Transcription Promotes Oncogene Induced Breast Cancer

The Mre11 complex (comprising Mre11, Rad50, Nbs1) is integral to the maintenance of genome stability. We previously showed that a hypomorphic Mre11 mutant mouse strain ( Mre11 ATLD1/ATLD1 ) was highly susceptible to oncogene induced breast cancer. Here we used a mammary organoid system to examine which Mre11 dependent responses are tumor suppressive. We found that Mre11 ATLD1/ATLD1 organoids exhibited an elevated interferon stimulated gene (ISG) signature and sustained changes in chromatin accessibility. This Mre11 ATLD1/ATLD1 phenotype depended on DNA binding of a nuclear innate immune sensor, IFI205. Ablation of Ifi205 in Mre11 ATLD1/ATLD1 organoids restored baseline and oncogene-induced chromatin accessibility patterns to those observed in WT . Implantation of Mre11 ATLD1/ATLD1 organoids and activation of oncogene led to aggressive metastatic breast cancer. This outcome was reversed in implanted Ifi205 -/- Mre11 ATLD1/ATLD1 organoids. These data reveal a connection between innate immune signaling and tumor suppression in mammary epithelium. Given the abundance of aberrant DNA structures that arise in the context of genome instability syndromes, the data further suggest that cancer predisposition in those contexts may be partially attributable to tonic innate immune transcriptional programs.

ctDNA-based detection of molecular residual disease in stage I-III non-small cell lung cancer patients treated with definitive radiotherapy

Background

Sensitive and reliable biomarkers for early detection of recurrence are needed to improve post-definitive radiation risk stratification, disease management, and outcomes for patients with unresectable early-stage or locally advanced non-small cell lung cancer (NSCLC) who are treated with definitive radiation therapy (RT). This prospective, multistate single-center, cohort study investigated the association of circulating tumor DNA (ctDNA) status with recurrence in patients with unresectable stage I-III NSCLC who underwent definitive RT.

Methods

A total of 70 serial plasma samples from 17 NSCLC patients were collected before, during, and after treatment. A personalized, tumor-informed ctDNA assay was used to track a set of up to 16 somatic, single nucleotide variants in the associated patient's plasma samples.

Results

Pre-treatment ctDNA detection rate was 82% (14/17) and varied based on histology and stage. ctDNA was detected in 35% (6/17) of patients at the first post-RT timepoint (median of 1.66 months following the completion of RT), all of whom subsequently developed clinical progression. At this first post-RT time point, patients with ctDNA-positivity had significantly worse progression-free survival (PFS) [hazard ratio (HR): 24.2, p=0.004], and ctDNA-positivity was the only significant prognostic factor associated with PFS (HR: 13.4, p=0.02) in a multivariate analysis. All patients who developed clinical recurrence had detectable ctDNA with an average lead time over radiographic progression of 5.4 months, and post-RT ctDNA positivity was significantly associated with poor PFS (p<0.0001).

Conclusion

Personalized, longitudinal ctDNA monitoring can detect recurrence early in patients with unresectable NSCLC patients undergoing curative radiation and potentially risk-stratify patients who might benefit most from treatment intensification.

Long-molecule scars of backup DNA repair in BRCA1- and BRCA2-deficient cancers

Homologous recombination (HR) deficiency is associated with DNA rearrangements and cytogenetic aberrations1. Paradoxically, the types of DNA rearrangements that are specifically associated with HR-deficient cancers only minimally affect chromosomal structure2. Here, to address this apparent contradiction, we combined genome-graph analysis of short-read whole-genome sequencing (WGS) profiles across thousands of tumours with deep linked-read WGS of 46 BRCA1- or BRCA2-mutant breast cancers. These data revealed a distinct class of HR-deficiency-enriched rearrangements called reciprocal pairs. Linked-read WGS showed that reciprocal pairs with identical rearrangement orientations gave rise to one of two distinct chromosomal outcomes, distinguishable only with long-molecule data. Whereas one (cis) outcome corresponded to the copying and pasting of a small segment to a distant site, a second (trans) outcome was a quasi-balanced translocation or multi-megabase inversion with substantial (10 kb) duplications at each junction. We propose an HR-independent replication-restart repair mechanism to explain the full spectrum of reciprocal pair outcomes. Linked-read WGS also identified single-strand annealing as a repair pathway that is specific to BRCA2 deficiency in human cancers. Integrating these features in a classifier improved discrimination between BRCA1- and BRCA2-deficient genomes. In conclusion, our data reveal classes of rearrangements that are specific to BRCA1 or BRCA2 deficiency as a source of cytogenetic aberrations in HR-deficient cells.

Genomic analysis and clinical correlations of non-small cell lung cancer brain metastasis

Up to 50% of patients with non-small cell lung cancer (NSCLC) develop brain metastasis (BM), yet the study of BM genomics has been limited by tissue access, incomplete clinical data, and a lack of comparison with paired extracranial specimens. Here we report a cohort of 233 patients with resected and sequenced (MSK-IMPACT) NSCLC BM and comprehensive clinical data. With matched samples (47 primary tumor, 42 extracranial metastatic), we show CDKN2A/B deletions and cell cycle pathway alterations to be enriched in the BM samples. Meaningful clinico-genomic correlations are noted, namely EGFR alterations in leptomeningeal disease (LMD) and MYC amplifications in multifocal regional brain progression. Patients who developed early LMD frequently have had uncommon, multiple, and persistently detectable EGFR driver mutations. The distinct mutational patterns identified in BM specimens compared to other tissue sites suggest specific biologic underpinnings of intracranial progression.

A Multiparameter Molecular Classifier to Predict Response to Neoadjuvant Lapatinib plus Trastuzumab without Chemotherapy in HER2+ Breast Cancer

PURPOSE

Clinical trials reported 25% to 30% pathologic complete response (pCR) rates in HER2+ patients with breast cancer treated with anti-HER2 therapies without chemotherapy. We hypothesize that a multiparameter classifier can identify patients with HER2-"addicted" tumors who may benefit from a chemotherapy-sparing strategy.

EXPERIMENTAL DESIGN

Baseline HER2+ breast cancer specimens from the TBCRC023 and PAMELA trials, which included neoadjuvant treatment with lapatinib and trastuzumab, were used. In the case of estrogen receptor-positive (ER+) tumors, endocrine therapy was also administered. HER2 protein and gene amplification (ratio), HER2-enriched (HER2-E), and PIK3CA mutation status were assessed by dual gene protein assay (GPA), research-based PAM50, and targeted DNA-sequencing. GPA cutoffs and classifier of response were constructed in TBCRC023 using a decision tree algorithm, then validated in PAMELA.

RESULTS

In TBCRC023, 72 breast cancer specimens had GPA, PAM50, and sequencing data, of which 15 had pCR. Recursive partitioning identified cutoffs of HER2 ratio ≥ 4.6 and %3+ IHC staining ≥ 97.5%. With PAM50 and sequencing data, the model added HER2-E and PIK3CA wild-type (WT). For clinical implementation, the classifier was locked as HER2 ratio ≥ 4.5, %3+ IHC staining ≥ 90%, and PIK3CA-WT and HER2-E, yielding 55% and 94% positive (PPV) and negative (NPV) predictive values, respectively. Independent validation using 44 PAMELA cases with all three biomarkers yielded 47% PPV and 82% NPV. Importantly, our classifier's high NPV signifies its strength in accurately identifying patients who may not be good candidates for treatment deescalation.

CONCLUSIONS

Our multiparameter classifier differentially identifies patients who may benefit from HER2-targeted therapy alone from those who need chemotherapy and predicts pCR to anti-HER2 therapy alone comparable with chemotherapy plus dual anti-HER2 therapy in unselected patients.

Non-cell-autonomous cancer progression from chromosomal instability

Chromosomal instability (CIN) is a driver of cancer metastasis1-4, yet the extent to which this effect depends on the immune system remains unknown. Using ContactTracing-a newly developed, validated and benchmarked tool to infer the nature and conditional dependence of cell-cell interactions from single-cell transcriptomic data-we show that CIN-induced chronic activation of the cGAS-STING pathway promotes downstream signal re-wiring in cancer cells, leading to a pro-metastatic tumour microenvironment. This re-wiring is manifested by type I interferon tachyphylaxis selectively downstream of STING and a corresponding increase in cancer cell-derived endoplasmic reticulum (ER) stress response. Reversal of CIN, depletion of cancer cell STING or inhibition of ER stress response signalling abrogates CIN-dependent effects on the tumour microenvironment and suppresses metastasis in immune competent, but not severely immune compromised, settings. Treatment with STING inhibitors reduces CIN-driven metastasis in melanoma, breast and colorectal cancers in a manner dependent on tumour cell-intrinsic STING. Finally, we show that CIN and pervasive cGAS activation in micronuclei are associated with ER stress signalling, immune suppression and metastasis in human triple-negative breast cancer, highlighting a viable strategy to identify and therapeutically intervene in tumours spurred by CIN-induced inflammation.

ESMO expert consensus statements (ECS) on the definition, diagnosis, and management of HER2-low breast cancer

Human epidermal growth factor receptor 2 (HER2)-low breast cancer has recently emerged as a targetable subset of breast tumors, based on the evidence from clinical trials of novel anti-HER2 antibody-drug conjugates. This evolution has raised several biological and clinical questions, warranting the establishment of consensus to optimally treat patients with HER2-low breast tumors. Between 2022 and 2023, the European Society for Medical Oncology (ESMO) held a virtual consensus-building process focused on HER2-low breast cancer. The consensus included a multidisciplinary panel of 32 leading experts in the management of breast cancer from nine different countries. The aim of the consensus was to develop statements on topics that are not covered in detail in the current ESMO Clinical Practice Guideline. The main topics identified for discussion were (i) biology of HER2-low breast cancer; (ii) pathologic diagnosis of HER2-low breast cancer; (iii) clinical management of HER2-low metastatic breast cancer; and (iv) clinical trial design for HER2-low breast cancer. The expert panel was divided into four working groups to address questions relating to one of the four topics outlined above. A review of the relevant scientific literature was conducted in advance. Consensus statements were developed by the working groups and then presented to the entire panel for further discussion and amendment before voting. This article presents the developed statements, including findings from the expert panel discussions, expert opinion, and a summary of evidence supporting each statement.

Spatial analyses of immune cell infiltration in cancer: current methods and future directions: A report of the International Immuno-Oncology Biomarker Working Group on Breast Cancer

Modern histologic imaging platforms coupled with machine learning methods have provided new opportunities to map the spatial distribution of immune cells in the tumor microenvironment. However, there exists no standardized method for describing or analyzing spatial immune cell data, and most reported spatial analyses are rudimentary. In this review, we provide an overview of two approaches for reporting and analyzing spatial data (raster versus vector-based). We then provide a compendium of spatial immune cell metrics that have been reported in the literature, summarizing prognostic associations in the context of a variety of cancers. We conclude by discussing two well-described clinical biomarkers, the breast cancer stromal tumor infiltrating lymphocytes score and the colon cancer Immunoscore, and describe investigative opportunities to improve clinical utility of these spatial biomarkers. © 2023 The Pathological Society of Great Britain and Ireland.

Clinicopathologic and genomic features of lobular like invasive mammary carcinoma: is it a distinct entity?

This study describes "lobular-like invasive mammary carcinomas" (LLIMCas), a group of low- to intermediate-grade invasive mammary carcinomas with discohesive, diffusely infiltrative cells showing retained circumferential membranous immunoreactivity for both E-cadherin and p120. We analyzed the clinical-pathologic features of 166 LLIMCas compared to 104 classical invasive lobular carcinomas (ILCs) and 100 grade 1 and 2 invasive ductal carcinomas (IDCs). Tumor size and pT stage of LLIMCas were intermediate between IDCs and ILCs, and yet often underestimated on imaging and showed frequent positive margins on the first resection. Despite histomorphologic similarities to classical ILC, the discohesion in LLIMCa was independent of E-cadherin/p120 immunophenotypic alteration. An exploratory, hypothesis-generating analysis of the genomic features of 14 randomly selected LLIMCas and classical ILCs (7 from each category) was performed utilizing an FDA-authorized targeted capture sequencing assay (MSK-IMPACT). None of the seven LLIMCas harbored CDH1 loss-of-function mutations, and none of the CDH1 alterations detected in two of the LLIMCas was pathogenic. In contrast, all seven ILCs harbored CDH1 loss-of-function mutations coupled with the loss of heterozygosity of the CDH1 wild-type allele. Four of the six evaluable LLIMCas were positive for CDH1 promoter methylation, which may partially explain the single-cell infiltrative morphology seen in LLIMCa. Further studies are warranted to better define the molecular basis of the discohesive cellular morphology in LLIMCa. Until more data becomes available, identifying LLIMCas and distinguishing them from typical IDCs and ILCs would be justified. In patients with LLIMCas, preoperative MRI should be entertained to guide surgical management.

Stem cell-like T cell depletion in the recurrent head and neck cancer immune microenvironment

Human papilloma virus (HPV)-related oncogenesis in head and neck cancer establishes a local microenvironment rich with immune cells, however the composition of the microenvironment in recurrent disease following definitive treatment is poorly understood. Here, we investigate the composition and spatial relationships between tumor and immune cells in recurrent head and neck cancer following curative intent chemoradiotherapy. Multiplexed immunofluorescence with 12 unique markers, through two multiplex immunofluorescent panels, was performed to evaluate 27 tumor samples including 18 pre-treatment primary and 9 paired recurrent tumors. Tumor and immune cell populations were phenotyped and quantified using a previously validated semi-automated digital pathology platform for cell segmentation. Spatial analysis was conducted by evaluating immune cells within the tumor, peri-tumoral stroma, and distant stroma. Initial tumors in patients with subsequent recurrence were found to be enriched in tumor associated macrophages and displayed an immune excluded spatial distribution. Recurrent tumors after chemoradiation were hypo-inflamed, with a statistically significant reduction in the recently identified stem-like TCF1+ CD8 T-cells, which normally function to maintain HPV-specific immune responses in the setting of chronic antigen exposure. Our findings indicate that the tumor microenvironment of recurrent HPV-related head and neck cancers displays a reduction in stem-like T cells, consistent with an immune microenvironment with a reduced ability to mount T-cell-driven anti-tumor immune responses.

Rare subtypes of triple negative breast cancer: Current understanding and future directions

Rare subtypes of triple-negative breast cancers (TNBC) are a heterogenous group of tumors, comprising 5-10% of all TNBCs. Despite accounting for an absolute number of cases in aggregate approaching that of other less common, but well studied solid tumors, rare subtypes of triple-negative disease remain understudied. Low prevalence, diagnostic challenges and overlapping diagnoses have hindered consistent categorization of these breast cancers. Here we review epidemiology, histology and clinical and molecular characteristics of metaplastic, triple-negative lobular, apocrine, adenoid cystic, secretory and high-grade neuroendocrine TNBCs. Medullary pattern invasive ductal carcinoma no special type, which until recently was a considered a distinct subtype, is also discussed. With this background, we review how applying biological principals often applied to study TNBC no special type could improve our understanding of rare TNBCs. These could include the utilization of targeted molecular approaches or disease agnostic tools such as tumor mutational burden or germline mutation-directed treatments. Burgeoning data also suggest that pathologic response to neoadjuvant therapy and circulating tumor DNA have value in understanding rare subtypes of TNBC. Finally, we discuss a framework for advancing disease-specific knowledge in this space. While the conduct of randomized trials in rare TNBC subtypes has been challenging, re-envisioning trial design and technologic tools may offer new opportunities. These include embedding rare TNBC subtypes in umbrella studies of rare tumors, retrospective review of contemporary trials, prospective identification of patients with rare TNBC subtypes entering on clinical trials and querying big data for outcomes of patients with rare breast tumors.

Overcoming the challenges to implementation of artificial intelligence in pathology

Pathologists worldwide are facing remarkable challenges with increasing workloads and lack of time to provide consistently high-quality patient care. The application of artificial intelligence (AI) to digital whole-slide images has the potential of democratizing the access to expert pathology and affordable biomarkers by supporting pathologists in the provision of timely and accurate diagnosis as well as supporting oncologists by directly extracting prognostic and predictive biomarkers from tissue slides. The long-awaited adoption of AI in pathology, however, has not materialized, and the transformation of pathology is happening at a much slower pace than that observed in other fields (eg, radiology). Here, we provide a critical summary of the developments in digital and computational pathology in the last 10 years, outline key hurdles and ways to overcome them, and provide a perspective for AI-supported precision oncology in the future.

Camonsertib in DNA damage response-deficient advanced solid tumors: phase 1 trial results

Predictive biomarkers of response are essential to effectively guide targeted cancer treatment. Ataxia telangiectasia and Rad3-related kinase inhibitors (ATRi) have been shown to be synthetic lethal with loss of function (LOF) of ataxia telangiectasia-mutated (ATM) kinase, and preclinical studies have identified ATRi-sensitizing alterations in other DNA damage response (DDR) genes. Here we report the results from module 1 of an ongoing phase 1 trial of the ATRi camonsertib (RP-3500) in 120 patients with advanced solid tumors harboring LOF alterations in DDR genes, predicted by chemogenomic CRISPR screens to sensitize tumors to ATRi. Primary objectives were to determine safety and propose a recommended phase 2 dose (RP2D). Secondary objectives were to assess preliminary anti-tumor activity, to characterize camonsertib pharmacokinetics and relationship with pharmacodynamic biomarkers and to evaluate methods for detecting ATRi-sensitizing biomarkers. Camonsertib was well tolerated; anemia was the most common drug-related toxicity (32% grade 3). Preliminary RP2D was 160 mg weekly on days 1-3. Overall clinical response, clinical benefit and molecular response rates across tumor and molecular subtypes in patients who received biologically effective doses of camonsertib (>100 mg d-1) were 13% (13/99), 43% (43/99) and 43% (27/63), respectively. Clinical benefit was highest in ovarian cancer, in tumors with biallelic LOF alterations and in patients with molecular responses. ClinicalTrials.gov registration: NCT04497116 .

NR4A3 Expression Is Consistently Absent in Acinic Cell Carcinomas of the Breast: A Potential Nosologic Shift

Acinic cell carcinoma (AciCC) is a tumor that is recognized in both the breast and salivary glands. Recently, the recurrent genomic rearrangement, t(4;9)(q13;q31) was identified in salivary AciCC that results in constitutive upregulation of the nuclear transcription factor NR4A3, which can be detected by immunohistochemistry. In this study, we sought to evaluate NR4A3 expression in breast AciCC using immunohistochemistry. Strong and diffuse nuclear staining was considered a positive result. Sixteen AciCCs were studied, including 8 pure AciCCs and 8 AciCCs admixed with other types (invasive carcinoma of no special type in 5 cases and metaplastic carcinoma in 3 cases). All 16 AciCCs showed negative results for NR4A3 expression. Four cases with available material were evaluated for rearrangements of the NR4A3 gene by fluorescence in situ hybridization and no rearrangements were observed. Whole-genome sequencing of 1 AciCC revealed a TP53 splice-site mutation, high levels of genomic instability, and genomic features of homologous recombination DNA repair defects; a structural variant analysis of this case did not reveal the presence of a t(4;9) rearrangement. We conclude that breast AciCCs consistently lack NR4A3 rearrangement or overexpression, unlike most of the salivary AciCCs, and that consistent with previous results, breast AciCCs are associated with genomic alterations more similar to those seen in triple-negative breast carcinomas than salivary gland AciCCs. These results suggest that unlike other salivary gland-like tumors that occur in the breast, the molecular underpinnings of the salivary gland and breast AciCCs are different and that the salivary gland and breast AciCCs likely represent distinct entities.

Dramatic, durable response to therapy in gBRCA2-mutated pancreas neuroendocrine carcinoma: opportunity and challenge

Poorly differentiated pancreatic neuroendocrine tumors (PDNEC), are a subtype of pancreatic cancer encompassing both small cell and large cell neuroendocrine carcinoma subtypes, and are characterized as distinct in terms of biology and prognosis compared to the more common pancreatic adenocarcinoma. Until recently, there has been a paucity of data on the genomic features of this cancer type. We describe a male patient diagnosed with PDNEC and extensive metastatic disease in the liver at diagnosis. Genomic analysis demonstrated a germline pathogenic variant in BRCA2 with somatic loss-of-heterozygosity of the BRCA2 wild-type allele. Following a favorable response to platinum-based chemotherapy (and the addition of immunotherapy), the patient received maintenance therapy with olaparib, which resulted in a further reduction on follow-up imaging (Fig. 1). After seventeen months of systemic control with olaparib, the patient developed symptomatic central nervous system metastases, which harboured a BRCA2 reversion mutation. No other sites of disease progression were observed. Herein, we report an exceptional outcome through the incorporation of a personalized management approach for a patient with a pancreatic PDNEC, guided by comprehensive genomic sequencing.

Biological insights and novel biomarker discovery through deep learning approaches in breast cancer histopathology

Breast cancer remains a highly prevalent disease with considerable inter- and intra-tumoral heterogeneity complicating prognostication and treatment decisions. The utilization and depth of genomic, transcriptomic and proteomic data for cancer has exploded over recent times and the addition of spatial context to this information, by understanding the correlating morphologic and spatial patterns of cells in tissue samples, has created an exciting frontier of research, histo-genomics. At the same time, deep learning (DL), a class of machine learning algorithms employing artificial neural networks, has rapidly progressed in the last decade with a confluence of technical developments - including the advent of modern graphic processing units (GPU), allowing efficient implementation of increasingly complex architectures at scale; advances in the theoretical and practical design of network architectures; and access to larger datasets for training - all leading to sweeping advances in image classification and object detection. In this review, we examine recent developments in the application of DL in breast cancer histology with particular emphasis of those producing biologic insights or novel biomarkers, spanning the extraction of genomic information to the use of stroma to predict cancer recurrence, with the aim of suggesting avenues for further advancing this exciting field.

Abstract 6421: Molecular profiles and single cell analysis identify immunogenic pancreatic ductal adenocarcinoma (iPDAC)

Most pancreatic ductal adenocarcinomas (PDAC) are lethal and resistant to immunotherapy. Thus, identifying the immunogenic subgroup (iPDAC) and therapeutic targets can save lives. Herein, we present molecular features of iPDAC. 3 cohorts (A, B, C) from 288 patients whose sequenced tumors (MSK-IMPACT) were classified by homologous recombination deficiency groups. MSI-H were excluded. Survival, tumor mutation burden, genomic instability score, and enriched pathways for each cohort are included in Table 1. Patients in A (BRCA1/2/PALB2) had longer survivals vs B/C. 61 samples were selected for bulk RNAseq analysis for A vs C. Gene Ontology was enriched for upregulated humoral, T cell, and neutrophil immunity. CIBERSORT suggested higher infiltration of gamma delta T (Tgd) cells (p=0.039) and neutrophils (p=0.012), but lower Treg (p=0.001). Multidimensional insights in cellular components of cancer, immune, stroma, and neural genes were obtained by single nuclear RNA (snRNAseq) analysis from 30 biopsies for A vs C. 10x Genomics Chromium platform for library and Scanpy for computational analysis after Cell Ranger pipelines were used. 61,868 nuclei were profiled from 18 (13 baseline and 5 matched longitudinal) samples after quality evaluation. UMAP accurately clustered cells from each patient. Long-term survivors (LTS) had heterogenous baseline immune cell infiltrates of plasma cells, neutrophils, and CD8 (+) cytotoxic T cells. In matched samples of LTS, evolution of more prominent CD8 (+) T cells, macrophage, plasma cell, and neutrophil were observed. Single nucleus T-Cell Receptor sequencing for clonal trajectory inference will be done to determine the associated single cell molecular features contributing to iPDAC and identify novel targets for future intervention.

Table 1. Cohort (Total: N=288) A: core HRD (BRCA1/2/PALB2) B: non-core HRD (ATM, BARD1, BLM, CHEK2, RAD50, RAD51C, RTEL1, MUTYH) C: others without HR-gene alterations Number (%) 48 (16.6) 19 (6.5) 221 (76) Median overall survival (95% confidence Interval) 33 months (3.6-64) 16 (11- not reached) 16 (14-18) Tumor Mutation Burden (TMB) 4.4 3.5 3.9 Genomic Instability Score (GIS, HRD score) 26 12 13 Gene Ongology term, enrichment score, adjusted p-value Adaptive immune response, GO:0002250, 0.49, 1.69e-10 Not included Reference to cohort A Humoral immune response, GO:0006959, 0.58, 1.67e-9 T cell activation, GO:0042110, 0.44, 2.75e-8 Neutrophil chemotaxis, GO:0030593, 0.73, 4.3e-10

Citation Format: Wungki Park, Catherine O'Connor, Shigeaki Umeda, Roshan Sharma, Yingjie Zhu, Elias-Ramzey Karnoub, Anna Varghese, Kevin C. Soares, Alejandro Jimemez, Asli Yavas, Kenneth H. Yu, Balachandran P. Vinod, Joanne F. Chou, Danny N. Khalil, Kelsen David, Hulya Sahin Ozkan, Olca Basturk, Marinela Capanu, Tal Nawy, Michael F. Berger, Ghassan K. Abou-Alfa, Jorge S. Reis-Filho, Ronan Chaligne, Nadeem Riaz, Dana Pe'er, Christine Iacobuzio-Donahue, Eileen M. O'Reilly. Molecular profiles and single cell analysis identify immunogenic pancreatic ductal adenocarcinoma (iPDAC) [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 6421.

Abstract 6063: Genomic characterization of non-small cell lung cancer (NSCLC) brain metastasis (BM) patients identifies novel alterations associated with tropisms and disease progression

Intro: Half of all patients with NSCLC develop BM during their clinical course. While modern NSCLC-directed agents yield excellent systemic response, most patients require focal BM treatment. Prior reports of BM genomics have been limited by low numbers and a lack of matched specimens. Here, we report the largest cohort to date of molecularly-profiled NSCLC BM samples with comprehensive clinical follow-up.

Methods: Clinical data and outcomes for 244 patients with NSCLC and resected BM were identified. Samples were assessed using MSK-IMPACT, a custom tumor-normal next generation sequencing assay. 51 (20.9%) patients had matched primary site tissue, and 44 (18%) patients had matched tissue from another metastatic site or CSF. Genomic alterations were filtered for driver variants using OncoKB. Publicly available genomic data for NSCLC primary samples was used for comparison against the primary samples from our BM cohort.

Results: The most frequently altered genes in BM tumors were TP53 (72%), CDKN2A (34%), KRAS (31%), KEAP1 (26%), and EGFR (21%). CDKN2A was more frequently altered in BM samples compared to NSCLC primary lesions (34% vs 14%, p = 0.003). Additionally, cell cycle pathway alterations were enriched in BM (56% vs 31%, p = 0.002). BM samples also had a significantly higher fraction of genome altered (FGA) relative to primary samples (p &lt; 0.0001). We then compared primary samples from BM patients against primary samples from metastatic NSCLC patients without BM and primary samples from non-metastatic NSCLC patients. We found an enrichment of alterations in TP53 (68.6% vs 27.7%, p &lt; 0.0001), NKX2-1 (11.4% vs 1.7%, p = 0.006), SMARCA4 (11.4% vs 2.1%, p = 0.01), RB1 (11.4% vs 1.7%, p = 0.006), and FOXA1 (11.4% vs 0.9%, p = 0.001) in the primary samples from BM patients compared to non-metastatic patients. Next, we grouped patients based on CNS progression patterns and found that EGFR alterations were enriched in patients with leptomeningeal progression when compared to patients without progression (42% vs 18%, p = 0.03).

Conclusions: In our cohort of molecularly-profiled NSCLC BM, we found an enrichment of cell cycle pathway alterations and a higher FGA in BMs compared to their primary tumor controls. Additionally, several genes were enriched in the primary tissue of patients that developed BM compared to primary tissue from non-metastatic patients. EGFR alterations were enriched in patients who develop leptomeningeal disease (LMD). Our work herein characterizes the genomic profiles of NSCLC patients with BM and identifies specific genes enriched in the primary tissue of BM patients compared to primary tissue from both non-BM metastatic patients and non-metastatic patients. Finally, our finding that EGFR alterations were enriched in patients with LMD suggests specific biologic underpinnings driving patterns of CNS progression.

Citation Format: Henry Walch, Anna Skakodub, Kathryn R. Tringale, Harish N. Vasudevan, Jordan Eichholz, Daniel W. Kelly, Emily Lebow, Nelson S. Moss, Kenny Kwok Hei Yu, Bob T. Li, Boris Mueller, Atif Khan, Yao Yu, Simon Powell, Jorge S. Reis-Filho, Brandon S. Imber, Pedram Razavi, Daniel R. Gomez, Nikolaus Schultz, Luke R. Pike. Genomic characterization of non-small cell lung cancer (NSCLC) brain metastasis (BM) patients identifies novel alterations associated with tropisms and disease progression [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 6063.

Abstract 5721: Automated annotation for large-scale clinicogenomic models of lung cancer treatment response and overall survival

The digitization of health records and prompt availability of tumor DNA sequencing results offer a chance to study the determinants of cancer outcomes with unprecedented richness; however, abstraction of key attributes from free text presents a major limitation to large-scale analyses. Using natural language processing (NLP), we derived sites of metastasis, prior treatment at outside institutions, programmed death ligand 1 (PD-L1) levels, and smoking status from records of patients with tumor sequencing to create a richly annotated clinicogenomic cohort. We sought to define whether combining features would improve models of overall survival (OS) and treatment response as validated in a multi-institution, manually curated cohort. We leveraged the manually curated AACR GENIE Biopharma Collaborative (BPC) dataset to train NLP algorithms to abstract the aforementioned features from overlapping records available at Memorial Sloan Kettering (MSK). All models achieved precision and recall &gt; 0.85. We deployed these algorithms to records of all MSK patients with non-small cell lung cancer (NSCLC) and tumor profiling with our FDA-authorized institutional targeted sequencing platform (N=7,015). These labels were combined with genomic, demographic, histopathologic, internal treatment and staging data to train random survival forests (RSF) to predict OS and time-to-next-treatment (TTNT) for molecularly targeted and immunotherapies. RSFs trained on the MSK NSCLC cohort were validated with the curated, non-MSK BPC NSCLC cohort (N=977). The addition of NLP-derived variables to genomic features enhanced RSF predictive power for OS (c-index, 10x bootstrap 95%CI: 0.58, 0.57-0.59 vs 0.75, 0.74-0.76 combined) and targeted and immunotherapy TTNT. The size of the MSK NSCLC cohort enabled discovery of associations between metastatic sites, PD-L1 status, genomics, and TTNTs not apparent in the smaller BPC cohort. We measured the added predictive value of variables not available in BPC with MSK-only cross-validation analyses. White blood cell differential counts and additional tissue genomic features including tumor mutational burden and fraction genome altered added minimally, while circulating tumor DNA sequencing added prognostic power for OS over other factors including disease burden

Using NLP we present a large NSCLC cohort with rich clinicoradiographic annotation, leading to superior models of patient outcomes. Our data uncovers associations not observed in smaller, manually curated cohorts and provides a foundation for further research in therapy choice and prognostication.

Citation Format: Justin Jee, Chris Fong, Karl Pichotta, Thinh Tran, Anisha Luthra, Mirella Altoe, Steven Maron, Ronglai Shen, Si-Yang Liu, Michele Waters, Joseph Kholodenko, Brooke Mastrogiacomo, Susie Kim, A Rose Brannon, Michael F. Berger, Axel Martin, Jason Chang, Anton Safonov, Jorge S. Reis-Filho, Deborah Schrag, Sohrab P. Shah, Pedram Razavi, Bob T. Li, Gregory J. Riely, Nikolaus Schultz. Automated annotation for large-scale clinicogenomic models of lung cancer treatment response and overall survival. [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 5721.

Molecular Biomarkers of Disease Outcomes and Mechanisms of Acquired Resistance to First-Line Osimertinib in Advanced EGFR-Mutant Lung Cancers

INTRODUCTION

Preferred first-line treatment for patients with metastatic EGFR-mutant lung cancer is osimertinib, yet it is not known whether patient outcomes may be improved by identifying and intervening on molecular markers associated with therapeutic resistance.

METHODS

All patients with metastatic EGFR-mutant lung cancer treated with first-line osimertinib at the Memorial Sloan Kettering Cancer Center (n = 327) were identified. Available pretreatment and postprogression tumor samples underwent targeted gene panel sequencing and mutational signature analysis using SigMA algorithm. Progression-free survival (PFS) and overall survival were estimated using the Kaplan-Meier method.

RESULTS

Using multivariate analysis, baseline atypical EGFR (median PFS = 5.8 mo, p < 0.001) and concurrent TP53/RB1 alterations (median PFS = 10.5 mo, p = 0.015) were associated with shorter PFS on first-line osimertinib. Of 95 patients with postprogression biopsies, acquired resistance mechanisms were identified in 52% (off-target, n = 24; histologic transformation, n = 14; on-target, n = 12), with MET amplification (n = 9), small cell lung transformation (n = 7), and acquired EGFR amplification (n = 7), the most frequently identified mechanisms. Although there was no difference in postprogression survival on the basis of identified resistance (p = 0.07), patients with subsequent second-line therapy tailored to postprogression biopsy results had improved postprogression survival (hazard ratio = 0.09, p = 0.006). The paired postprogression tumors had higher tumor mutational burden (p = 0.008) and further dominant APOBEC mutational signatures (p = 0.07) compared with the pretreatment samples.

CONCLUSIONS

Patients with EGFR-mutant lung cancer treated with first-line osimertinib have improved survival with treatment adaptation on the basis of identified mechanisms of resistance at time of progression using tissue-based genomic analysis. Further survival gains may be achieved using risk-based treatment adaptation of pretreatment genomic alterations.

Impact of immune infiltration signatures on prognosis in endometrial carcinoma is dependent on the underlying molecular subtype

OBJECTIVES

Increased numbers of tumor infiltrating lymphocytes (TIL) in endometrial cancer (EC) are associated with improved survival, but it is unclear how this prognostic significance relates to the underlying EC molecular subtype. In this explorative hypothesis-generating study, we sought to define the immune signatures associated with the molecular subtypes of EC (i.e., POLE-mutated, microsatellite unstable (MSI-high), copy number (CN)-low, and CN-high) and to determine their correlation with patient outcomes.

METHODS

RNA-sequencing and molecular subtype data of 232 primary ECs were obtained from The Cancer Genome Atlas. Deconvolution of bulk gene expression data was performed using single sample Gene Set Enrichment Analysis (ssGSEA) and Cell type Identification By Estimating Relative Subsets Of known RNA Transcripts (CIBERSORT). The association of the resultant immune signatures with overall survival was determined across molecular subtypes.

RESULTS

Statistically significant differences in enrichment were identified in 16/30 and 6/23 immune gene sets by ssGSEA and CIBERSORT, respectively. Signature of CD8+ cells in ECs of CN-high molecular subtype was associated with improved overall survival by ssGSEA (p = 0.0108), while CD8 signatures did not appear to be prognostic in MSI-high (p = 0.74) or CN-low EC molecular subtypes (p = 0.793). Of all molecular subtypes, CN-high ECs exhibited the lowest levels of CD8+ T cell infiltration. Consistent with antigen-induced T cell activation and exhaustion, enrichment for immunomodulatory receptors was predominantly observed in ECs of MSI-high and POLE-mutated molecular subtypes.

CONCLUSIONS

Deconvolution of bulk gene expression data can be used to identify populations of immune infiltrated endometrial cancers with improved survival. These data support the existence of unique mechanisms of immune resistance within molecular subgroups of the disease.

Detection of biallelic loss of DNA repair genes in formalin-fixed, paraffin-embedded tumor samples using a novel tumor-only sequencing panel

Patient selection for synthetic lethal (SL)-based cancer therapy may be improved by assessment of gene-specific loss of heterozygosity (LOH) and biallelic loss of function (LOF). This report describes SyNthetic lethal Interactions for Precision Diagnostics (SNiPDx), a targeted next-generation sequencing (NGS) panel for detection of LOH and biallelic LOF alterations in 26 target genes focused on DNA damage response (DDR) pathways, in tumor-only formalin-fixed, paraffin-embedded (FFPE) samples. NGS was performed across all exons of these 26 genes and encompassed a total of 7632 genome-wide single nucleotide polymorphisms (SNPs) on genomic DNA from 80 FFPE solid tumor samples. The Fraction and Allele-Specific Copy Number Estimates from Tumor Sequencing (FACETS) algorithm was optimized to assess tumor purity and copy number based on heterozygous SNPs. SNiPDx demonstrated high sensitivity (95%) and specificity (91%) for LOH detection compared with whole genome sequencing. Positive agreement with local NGS-based testing in the detection of genetic alterations was 95%. SNiPDx detected 93% of biallelic ATM LOF mutations, 100% of ATM single nucleotide variants and small insertion/deletions, and 100% of all ATM LOH status events identified by orthogonal NGS-based testing. SNiPDx is a novel, clinically feasible test for analysis of allelic status in FFPE tumor samples, which demonstrated high accuracy when compared with other NGS-based approaches in clinical use.

Direct prediction of Homologous Recombination Deficiency from routine histology in ten different tumor types with attention-based Multiple Instance Learning: a development and validation study

Background

Homologous Recombination Deficiency (HRD) is a pan-cancer predictive biomarker that identifies patients who benefit from therapy with PARP inhibitors (PARPi). However, testing for HRD is highly complex. Here, we investigated whether Deep Learning can predict HRD status solely based on routine Hematoxylin & Eosin (H&E) histology images in ten cancer types.

Methods

We developed a fully automated deep learning pipeline with attention-weighted multiple instance learning (attMIL) to predict HRD status from histology images. A combined genomic scar HRD score, which integrated loss of heterozygosity (LOH), telomeric allelic imbalance (TAI) and large-scale state transitions (LST) was calculated from whole genome sequencing data for n=4,565 patients from two independent cohorts. The primary statistical endpoint was the Area Under the Receiver Operating Characteristic curve (AUROC) for the prediction of genomic scar HRD with a clinically used cutoff value.

Results

We found that HRD status is predictable in tumors of the endometrium, pancreas and lung, reaching cross-validated AUROCs of 0.79, 0.58 and 0.66. Predictions generalized well to an external cohort with AUROCs of 0.93, 0.81 and 0.73 respectively. Additionally, an HRD classifier trained on breast cancer yielded an AUROC of 0.78 in internal validation and was able to predict HRD in endometrial, prostate and pancreatic cancer with AUROCs of 0.87, 0.84 and 0.67 indicating a shared HRD-like phenotype is across tumor entities.

Conclusion

In this study, we show that HRD is directly predictable from H&E slides using attMIL within and across ten different tumor types.

Clinical Trials and Digital Pathology-Toward Quantitative Therapeutic Immunohistochemistry and Tissue Hybridization

This Viewpoint discusses immunohistochemistry and tissue-hybridization-based diagnostics delivery and compares it with digital pathology.

High-Sensitivity Mutation Analysis of Cell-Free DNA for Disease Monitoring in Endometrial Cancer

PURPOSE

We sought to determine whether sequencing analysis of circulating cell-free DNA (cfDNA) in patients with prospectively accrued endometrial cancer captures the mutational repertoire of the primary lesion and allows for disease monitoring.

EXPERIMENTAL DESIGN

Peripheral blood was prospectively collected from 44 newly diagnosed patients with endometrial cancer over a 24-month period (i.e., baseline, postsurgery, every 6 months after). DNA from the primary endometrial cancers was subjected to targeted next-generation sequencing (NGS) of 468 cancer-related genes, and cfDNA to a high-depth NGS assay of 129 genes with molecular barcoding. Sequencing data were analyzed using validated bioinformatics methods.

RESULTS

cfDNA levels correlated with surgical stage in endometrial cancers, with higher levels of cfDNA being present in advanced-stage disease. Mutations in cfDNA at baseline were detected preoperatively in 8 of 36 (22%) patients with sequencing data, all of whom were diagnosed with advanced-stage disease, high tumor volume, and/or aggressive histologic type. Of the 38 somatic mutations identified in the primary tumors also present in the cfDNA assay, 35 (92%) and 38 (100%) were detected at baseline and follow-up, respectively. In 6 patients with recurrent disease, changes in circulating tumor DNA (ctDNA) fraction/variant allele fractions in cfDNA during follow-up closely mirrored disease progression and therapy response, with a lead time over clinically detected recurrence in two cases. The presence of ctDNA at baseline (P < 0.001) or postsurgery (P = 0.014) was significantly associated with reduced progression-free survival.

CONCLUSIONS

cfDNA sequencing analysis in patients with endometrial cancer at diagnosis has prognostic value, and serial postsurgery cfDNA analysis enables disease and treatment response monitoring. See related commentary by Grant et al., p. 305.

Molecular Characterization of Acquired Resistance to KRASG12C-EGFR Inhibition in Colorectal Cancer

With the combination of KRASG12C and EGFR inhibitors, KRAS is becoming a druggable target in colorectal cancer. However, secondary resistance limits its efficacy. Using cell lines, patient-derived xenografts, and patient samples, we detected a heterogeneous pattern of putative resistance alterations expected primarily to prevent inhibition of ERK signaling by drugs at progression. Serial analysis of patient blood samples on treatment demonstrates that most of these alterations are detected at a low frequency except for KRASG12C amplification, a recurrent resistance mechanism that rises in step with clinical progression. Upon drug withdrawal, resistant cells with KRASG12C amplification undergo oncogene-induced senescence, and progressing patients experience a rapid fall in levels of this alteration in circulating DNA. In this new state, drug resumption is ineffective as mTOR signaling is elevated. However, our work exposes a potential therapeutic vulnerability, whereby therapies that target the senescence response may overcome acquired resistance.

SIGNIFICANCE

Clinical resistance to KRASG12C-EGFR inhibition primarily prevents suppression of ERK signaling. Most resistance mechanisms are subclonal, whereas KRASG12C amplification rises over time to drive a higher portion of resistance. This recurrent resistance mechanism leads to oncogene-induced senescence upon drug withdrawal and creates a potential vulnerability to senolytic approaches. This article is highlighted in the In This Issue feature, p. 1.

Expanded genetic testing of GIST patients identifies high proportion of non-syndromic patients with germline alterations

Traditional genetic testing for patients with gastrointestinal stromal tumors (GISTs) focus on those with syndromic features. To assess whether expanded genetic testing of GIST patients could identify hereditary cancer predisposition, we analyzed matched tumor-germline sequencing results from 103 patients with GISTs over a 6-year period. Germline pathogenic/likely pathogenic (P/LP) variants in GIST-associated genes (SDHA, SDHB, SDHC, NF1, KIT) were identified in 69% of patients with KIT/PDGFRA-wildtype GISTs, 63% of whom did not have any personal or family history of syndromic features. To evaluate the frequency of somatic versus germline variants identified in tumor-only sequencing of GISTs, we analyzed 499 de-identified tumor-normal pairs. P/LP variants in certain genes (e.g., BRCA1/2, SDHB) identified in tumor-only sequencing of GISTs were almost exclusively germline in origin. Our results provide guidance for genetic testing of GIST patients and indicate that germline testing should be offered to all patients with KIT/PDGFRA-wildtype GISTs regardless of their history of syndromic features.

Comprehensive Genomic Profiling of Cell-Free Circulating Tumor DNA Detects Response to Ribociclib Plus Letrozole in a Patient with Metastatic Breast Cancer

Analysis of cell-free circulating tumor DNA obtained by liquid biopsy is a non-invasive approach that may provide clinically actionable information when conventional tissue biopsy is inaccessible or infeasible. Here, we followed a patient with hormone receptor-positive and human epidermal growth factor receptor (HER) 2-negative breast cancer who developed bone metastases seven years after mastectomy. We analyzed circulating cell-free DNA (cfDNA) extracted from plasma using high-depth massively parallel sequencing targeting 468 cancer-associated genes, and we identified a clonal hotspot missense mutation in the PIK3CA gene (3:178952085, A > G, H1047R) and amplification of the CCND1 gene. Whole-exome sequencing revealed that both alterations were present in the primary tumor. After treatment with ribociclib plus letrozole, the genetic abnormalities were no longer detected in cfDNA. These results underscore the clinical utility of combining liquid biopsy and comprehensive genomic profiling to monitor treatment response in patients with metastasized breast cancer.

APOBEC mutagenesis, kataegis, chromothripsis in EGFR-mutant osimertinib-resistant lung adenocarcinomas

BACKGROUND

Studies of targeted therapy resistance in lung cancer have primarily focused on single-gene alterations. Based on prior work implicating apolipoprotein b mRNA-editing enzyme, catalytic polypeptide-like (APOBEC) mutagenesis in histological transformation of epidermal growth factor receptor (EGFR)-mutant lung cancers, we hypothesized that mutational signature analysis may help elucidate acquired resistance to targeted therapies.

PATIENTS AND METHODS

APOBEC mutational signatures derived from an Food and Drug Administration-cleared multigene panel [Memorial Sloan Kettering Cancer Center Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT)] using the Signature Multivariate Analysis (SigMA) algorithm were validated against the gold standard of mutational signatures derived from whole-exome sequencing. Mutational signatures were decomposed in 3276 unique lung adenocarcinomas (LUADs), including 93 paired osimertinib-naïve and -resistant EGFR-mutant tumors. Associations between APOBEC and mechanisms of resistance to osimertinib were investigated. Whole-genome sequencing was carried out on available EGFR-mutant lung cancer samples (10 paired, 17 unpaired) to investigate large-scale genomic alterations potentially contributing to osimertinib resistance.

RESULTS

APOBEC mutational signatures were more frequent in receptor tyrosine kinase (RTK)-driven lung cancers (EGFR, ALK, RET, and ROS1; 25%) compared to LUADs at large (20%, P < 0.001); across all subtypes, APOBEC mutational signatures were enriched in subclonal mutations (P < 0.001). In EGFR-mutant lung cancers, osimertinib-resistant samples more frequently displayed an APOBEC-dominant mutational signature compared to osimertinib-naïve samples (28% versus 14%, P = 0.03). Specifically, mutations detected in osimertinib-resistant tumors but not in pre-treatment samples significantly more frequently displayed an APOBEC-dominant mutational signature (44% versus 23%, P < 0.001). EGFR-mutant samples with APOBEC-dominant signatures had enrichment of large-scale genomic rearrangements (P = 0.01) and kataegis (P = 0.03) in areas of APOBEC mutagenesis.

CONCLUSIONS

APOBEC mutational signatures are frequent in RTK-driven LUADs and increase under the selective pressure of osimertinib in EGFR-mutant lung cancer. APOBEC mutational signature enrichment in subclonal mutations, private mutations acquired after osimertinib treatment, and areas of large-scale genomic rearrangements highlights a potentially fundamental role for APOBEC mutagenesis in the development of resistance to targeted therapies, which may be potentially exploited to overcome such resistance.