No abstract

High-grade serous ovarian cancer (HGSOC) is an archetypal cancer of genomic instability1–4 patterned by distinct mutational processes5,6, tumour heterogeneity7–9 and intraperitoneal spread7,8,10. Immunotherapies have had limited efficacy in HGSOC11–13, highlighting an unmet need to assess how mutational processes and the anatomical sites of tumour foci determine the immunological states of the tumour microenvironment. Here we carried out an integrative analysis of whole-genome sequencing, single-cell RNA sequencing, digital histopathology and multiplexed immunofluorescence of 160 tumour sites from 42 treatment-naive patients with HGSOC. Homologous recombination-deficient HRD-Dup (BRCA1 mutant-like) and HRD-Del (BRCA2 mutant-like) tumours harboured inflammatory signalling and ongoing immunoediting, reflected in loss of HLA diversity and tumour infiltration with highly differentiated dysfunctional CD8+ T cells. By contrast, foldback-inversion-bearing tumours exhibited elevated immunosuppressive TGFβ signalling and immune exclusion, with predominantly naive/stem-like and memory T cells. Phenotypic state associations were specific to anatomical sites, highlighting compositional, topological and functional differences between adnexal tumours and distal peritoneal foci. Our findings implicate anatomical sites and mutational processes as determinants of evolutionary phenotypic divergence and immune resistance mechanisms in HGSOC. Our study provides a multi-omic cellular phenotype data substrate from which to develop and interpret future personalized immunotherapeutic approaches and early detection research.

Immunotherapy is used to treat almost all patients with advanced non-small cell lung cancer (NSCLC); however, identifying robust predictive biomarkers remains challenging. Here we show the predictive capacity of integrating medical imaging, histopathologic and genomic features to predict immunotherapy response using a cohort of 247 patients with advanced NSCLC with multimodal baseline data obtained during diagnostic clinical workup, including computed tomography scan images, digitized programmed death ligand-1 immunohistochemistry slides and known outcomes to immunotherapy. Using domain expert annotations, we developed a computational workflow to extract patient-level features and used a machine-learning approach to integrate multimodal features into a risk prediction model. Our multimodal model (area under the curve (AUC) = 0.80, 95% confidence interval (CI) 0.74–0.86) outperformed unimodal measures, including tumor mutational burden (AUC = 0.61, 95% CI 0.52–0.70) and programmed death ligand-1 immunohistochemistry score (AUC = 0.73, 95% CI 0.65–0.81). Our study therefore provides a quantitative rationale for using multimodal features to improve prediction of immunotherapy response in patients with NSCLC using expert-guided machine learning.

Objective Broad adoption of digital pathology (DP) is still lacking, and examples for DP connecting diagnostic, research, and educational use cases are missing. We blueprint a holistic DP solution at a large academic medical center ubiquitously integrated into clinical workflows; researchapplications including molecular, genetic, and tissue databases; and educational processes. Materials and Methods We built a vendor-agnostic, integrated viewer for reviewing, annotating, sharing, and quality assurance of digital slides in a clinical or research context. It is the first homegrown viewer cleared by New York State provisional approval in 2020 for primary diagnosis and remote sign-out during the COVID-19 (coronavirus disease 2019) pandemic. We further introduce an interconnected Honest Broker for BioInformatics Technology (HoBBIT) to systematically compile and share large-scale DP research datasets including anonymized images, redacted pathology reports, and clinical data of patients with consent. Results The solution has been operationally used over 3 years by 926 pathologists and researchers evaluating 288 903 digital slides. A total of 51% of these were reviewed within 1 month after scanning. Seamless integration of the viewer into 4 hospital systems clearly increases the adoption of DP. HoBBIT directly impacts the translation of knowledge in pathology into effective new health measures, including artificial intelligence–driven detection models for prostate cancer, basal cell carcinoma, and breast cancer metastases, developed and validated on thousands of cases. Conclusions We highlight major challenges and lessons learned when going digital to provide orientation for other pathologists. Building interconnected solutions will not only increase adoption of DP, but also facilitate next-generation computational pathology at scale for enhanced cancer research.

High-grade serous ovarian cancer (HGSOC) is an archetypal cancer of genomic instability patterned by distinct mutational processes, intratumoral heterogeneity and intraperitoneal spread. We investigated determinants of immune recognition and evasion in HGSOC to elucidate co-evolutionary processes underlying malignant progression and tumor immunity. Mutational processes and anatomic sites of tumor foci were key determinants of tumor microenvironment cellular phenotypes, inferred from whole genome sequencing, single-cell RNA sequencing, digital histopathology and multiplexed immunofluorescence of 160 tumor sites from 42 treatment-naive HGSOC patients. Homologous recombination-deficient (HRD)-Dup (BRCA1 mutant-like) and HRD-Del (BRCA2 mutant-like) tumors harbored increased neoantigen burden, inflammatory signaling and ongoing immunoediting, reflected in loss of HLA diversity and tumor infiltration with highly-differentiated dysfunctional CD8+ T cells. Foldback inversion (FBI, non-HRD) tumors exhibited elevated TGFβ signaling and immune exclusion, with predominantly naive/stem-like and memory T cells. Our findings implicate distinct immune resistance mechanisms across HGSOC subtypes which can inform future immunotherapeutic strategies.

No abstract

High-grade serous ovarian cancer (HGSOC) is an archetypal cancer of genomic instability patterned by distinct mutational processes, a high degree of tumor heterogeneity and intraperitoneal spread. As immunotherapies have thus far proven ineffective in this disease, we sought to establish the determinants of immune recognition, avoidance and evasion in disease natural history to gain insight into the co-evolutionary processes underlying malignant progression and host immunity. Accordingly we linked mutational processes and anatomic sites of tumor foci as determinants of tumor microenvironment (TME) cellular phenotypes within and between patients using genome-based stratification of homologous recombination proficient (HRP) and deficient (HRD) disease subtypes, and profiling single cell phenotypes from ~1 million cells including cancer cells, T cells, myeloid cells and fibroblasts derived from single cell RNA sequencing, and in situ spatial profiling of histopathology, cancer cell, T cell and macrophage states of 160 tumor sites obtained from 42 treatment-naive patients. Mutational processes in HRD-Dup (BRCA1 mutant-like) tumors were associated with cancer cell-intrinsic JAK/STAT signaling and predominance of highly-differentiated dysfunctional CD8+ T cells in the TME; HRD-Del (BRCA2 mutant-like) tumors were associated with cancer cell-intrinsic NF-κB and TNFα signaling and expansion of M2-type macrophages; and foldback inversion (FBI, HRP) tumors were associated with cancer cell-intrinsic TGFβ signaling and overall immune exclusion, with a predominance of naive/central memory-like T cells. Increased neoantigen burden and HLA loss of heterozygosity (LOH) were defining genomic features of the HRD, but not FBI tumors. These mechanisms of escape from immune predation, with distinct signalling activity and losses of HLA allelic diversity in HRD tumors, connect evolutionary selection with immunological phenotypic states. Multi-region sampling revealed substantial spatial variation, highlighting site-specific properties of the ovary and fallopian tube as putative “immune-privileged” sites. These results establish that in patients with widespread intraperitoneal disease, the local properties of organ sites may determine malignant cell selection and immune pruning. Furthermore, we observed that spatial cellular topology is a major determinant of tumor-immune interactions by in situ protein measurements, revealing ubiquitous PD1-PDL1 interactions in HRD tumors. Together, our findings yield mechanistic insights for how distinct mutational processes in HGSOC lead to diverse patterns of within- and between- patient variation in immune resistance, which can be exploited to optimize future immuno-therapeutic treatment strategies. Citation Format: Ignacio Vázquez-García, Florian Uhlitz, Nicholas Ceglia, Jamie L. Lim, Michelle Wu, Neeman Mohibullah, Arvin Eric B. Ruiz, Kevin M. Boehm, Viktoria Bojilova, Christopher J. Fong, Tyler Funnell, Diljot Grewal, Eliyahu Havasov, Samantha Leung, Arfath Pasha, Druv M. Patel, Maryam Pourmaleki, Nicole Rusk, Hongyu Shi, Rami Vanguri, Marc J. Williams, Allen W. Zhang, Vance Broach, Dennis S. Chi, Arnaud Da Cruz Paula, Ginger J. Gardner, Sarah H. Kim, Matthew Lennon, Kara Long Roche, Yukio Sonoda, Oliver Zivanovic, Ritika Kundra, Agnes Viale, Yonina Bykov, Fatemeh N. Derakhshan, Luke Geneslaw, Ana Maroldi, Andrea Schietinger, Travis J. Hollmann, Samuel F. Bakhoum, Robert A. Soslow, Lora H. Ellenson, Nadeem Abu-Rustum, Carol Aghajanian, Claire F. Friedman, Andrew McPherson, Britta Weigelt, MSK SPECTRUM Consortium, Dmitriy Zamarin, Sohrab P. Shah. Immune and malignant cell phenotypes of ovarian cancer are determined by distinct mutational processes [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 2553.

9064 Background: Immunotherapy is now given to almost all patients with advanced non-small cell lung cancer (NSCLC). However, developing robust biomarkers to predict benefit remains challenging. We set out to evaluate the predictive capacity of integrating medical imaging, histopathologic, and genomic features to develop a multimodal biomarker for immunotherapy response. Methods: We used baseline data that is routinely obtained during diagnostic clinical workup at a single center in patients with NSCLC and known outcomes following immunotherapy. The multimodal dataset included DNA alterations from NGS, CT scan images, and digitized PD-L1 immunohistochemistry (IHC) slides. Guided by experts in each field, we developed a workflow to extract data for each patient and used an attention-gated machine learning approach to integrate the features into a risk prediction model. Results: Our cohort included 247 patients with advanced NSCLC who received immunotherapy and had complete radiology, pathology, genomic, and clinical data. The patient cohort was 54% female, had a median age of 68 years (range 38-93), and 88% patients had a smoking history. Responders (CR/PR) vs non-responders (SD/PD) showed a median PFS and OS of 2.7 months (95% CI 2.5-3.0) and 11.4 months (95% CI 10.3-12.8), respectively. Of all patients, 187 (76%) had segmentable disease on chest CT scans. We used a radiomics approach and aggregated the average individual lesion predictions to construct patient-level response predictions which resulted in an overall AUC = 0.65, 95% CI 0.57-0.73. We next studied digitized FFPE slides of pre-treatment PD-L1 IHC staining of tumor specimens. Overall, 52% of slides showed PD-L1 tumor proportion score (TPS) ≥ 1% and were used to extract IHC-texture, a novel spatial characterization of PD-L1 staining. Logistic regression modeling on IHC-texture resulted in prediction accuracy of AUC = 0.62 (95% CI 0.51-0.73) which was inferior to the pathologist-assessed PD-L1 TPS (AUC = 0.73, 95% CI 0.65-0.81). We next implemented a dynamic deep attention-based multiple instance learning model with masking to evaluate the impact of combining features from all modalities. Our multimodal model (AUC = 0.80, 95% CI 0.74-0.86) outperformed unimodal measures, including tumor mutational burden (AUC = 0.61, 95% CI 0.52-0.70) and PD-L1 TPS (AUC = 0.73, 95% CI 0.65-0.81). Conclusions: Our study is a proof of concept for using multimodal features to improve prediction of immunotherapy response over standard-of-care approaches in patients with NSCLC using expert-guided machine learning.