Search

Abstract Immunohistochemistry (IHC) is the common companion diagnostics in targeted therapies. However, quantifying protein expressions in IHC images present a significant challenge, due to variability in manual scoring and inherent subjective interpretation. Deep learning (DL) offers a promising approach to address these issues, though current models require extensive training for each cancer and IHC type, limiting the practical application. We developed a Universal IHC (UIHC) analyzer, a DL-based tool that quantifies protein expression across different cancers and IHC types. This multi-cohort trained model outperformed conventional single-cohort models in analyzing unseen IHC images (Kappa score 0.578 vs. up to 0.509) and demonstrated consistent performance across varying positive staining cutoff values. In a discovery application, the UIHC model assigned higher tumor proportion scores to MET amplification cases, but not MET exon 14 splicing or other non-small cell lung cancer cases. This UIHC model represents a novel role for DL that further advances quantitative analysis of IHC.

Abstract Background Accurate classification of breast cancer molecular subtypes is crucial in determining treatment strategies and predicting clinical outcomes. This classification largely depends on the assessment of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) status. However, variability in interpretation among pathologists pose challenges to the accuracy of this classification. This study evaluates the role of artificial intelligence (AI) in enhancing the consistency of these evaluations. Methods AI-powered HER2 and ER/PR analyzers, consisting of cell and tissue models, were developed using 1,259 HER2, 744 ER, and 466 PR-stained immunohistochemistry (IHC) whole-slide images of breast cancer. External validation cohort comprising HER2, ER, and PR IHCs of 201 breast cancer cases were analyzed with these AI-powered analyzers. Three board-certified pathologists independently assessed these cases without AI annotation. Then, cases with differing interpretations between pathologists and the AI analyzer were revisited with AI assistance, focusing on evaluating the influence of AI assistance on the concordance among pathologists during the revised evaluation compared to the initial assessment. Results Reevaluation was required in 61 (30.3%), 42 (20.9%), and 80 (39.8%) of HER2, in 15 (7.5%), 17 (8.5%), and 11 (5.5%) of ER, and in 26 (12.9%), 24 (11.9%), and 28 (13.9%) of PR evaluations by the pathologists, respectively. Compared to initial interpretations, the assistance of AI led to a notable increase in the agreement among three pathologists on the status of HER2 (from 49.3 to 74.1%, p

Background The inflamed immune phenotype (IIP), defined by enrichment of tumor-infiltrating lymphocytes (TILs) within intratumoral areas, is a promising tumor-agnostic biomarker of response to immune checkpoint inhibitor (ICI) therapy. However, it is challenging to define the IIP in an objective and reproducible manner during manual histopathologic examination. Here, we investigate artificial intelligence (AI)-based immune phenotypes capable of predicting ICI clinical outcomes in multiple solid tumor types.Methods Lunit SCOPE IO is a deep learning model which determines the immune phenotype of the tumor microenvironment based on TIL analysis. We evaluated the correlation between the IIP and ICI treatment outcomes in terms of objective response rates (ORR), progression-free survival (PFS), and overall survival (OS) in a cohort of 1,806 ICI-treated patients representing over 27 solid tumor types retrospectively collected from multiple institutions.Results We observed an overall IIP prevalence of 35.2% and significantly more favorable ORRs (26.3% vs 15.8%), PFS (median 5.3 vs 3.1 months, HR 0.68, 95% CI 0.61 to 0.76), and OS (median 25.3 vs 13.6 months, HR 0.66, 95% CI 0.57 to 0.75) after ICI therapy in IIP compared with non-IIP patients, respectively (p

Comprehensive molecular profiles are required to understand and treat sarcomas, which comprise more than 70 different subtypes. Here, the authors profile the genomic landscape of 7494 sarcomas across 44 histologies using targeted panel sequencing and identify potential therapeutic targets.

Molecularly targeted cancer therapy has improved outcomes for patients with cancer with targetable oncoproteins, such as mutant EGFR in lung cancer. Yet, the long-term survival of these patients remains limited, because treatment responses are typically incomplete. One potential explanation for the lack of complete and durable responses is that oncogene-driven cancers with activating mutations of EGFR often harbor additional co-occurring genetic alterations. This hypothesis remains untested for most genetic alterations that co-occur with mutant EGFR. Here, we report the functional impact of inactivating genetic alterations of the mRNA splicing factor RNA-binding motif 10 (RBM10) that co-occur with mutant EGFR. RBM10 deficiency decreased EGFR inhibitor efficacy in patient-derived EGFR-mutant tumor models. RBM10 modulated mRNA alternative splicing of the mitochondrial apoptotic regulator Bcl-x to regulate tumor cell apoptosis during treatment. Genetic inactivation of RBM10 diminished EGFR inhibitor–mediated apoptosis by decreasing the ratio of (proapoptotic) Bcl-xS to (antiapoptotic) Bcl-xL isoforms of Bcl-x. RBM10 deficiency was a biomarker of poor response to EGFR inhibitor treatment in clinical samples. Coinhibition of Bcl-xL and mutant EGFR overcame the resistance induced by RBM10 deficiency. This study sheds light on the role of co-occurring genetic alterations and on the effect of splicing factor deficiency on the modulation of sensitivity to targeted kinase inhibitor cancer therapy.

An EGFR mutant with kinase domain duplication (EGFR-KDD) was previously identified in an index patient, but the functional and therapeutic implications remain unclear. Here, the authors show that KDD occurs in other ErbB receptors in multiple cancers, and characterize the mechanism and inhibition of EGFR-KDD.

Abstract Background Desmoplastic small round cell tumor (DSRCT) is a rare, highly aggressive, translocation-associated soft-tissue sarcoma that primarily affects children, adolescents, and young adults, with a striking male predominance. It is characterized by t(11;22) generating a novel EWSR1-WT1 fusion gene. Secondary genomic alterations are rarely described. Methods Tumor tissue from 83 DSRCT patients was assayed by hybrid-capture based comprehensive genomic profiling, FoundationOne® Heme next generation sequencing analysis of 406 genes and RNA sequencing of 265 genes. Tumor mutation burden was calculated from a minimum of 1.4 Mb sequenced DNA. Microsatellite instability status was determined by a novel algorithm analyzing 114 specific loci. Results Comprehensive genomic profiling identified several genomically-defined DSRCT subgroups. Recurrent genomic alterations were most frequently detected in FGFR4, ARID1A, TP53, MSH3, and MLL3 genes. With the exception of FGFR4, where the genomic alterations predicted activation, most of the alterations in the remaining genes predicted gene inactivation. No DSRCT were TMB or MSI high. Conclusions In summary, recurrent secondary somatic alterations in FGFR4, ARID1A, TP53, MSH3, and MLL3 were detected in 82% of DSRCT, which is significantly greater than previously reported. These alterations may have both prognostic and therapeutic implications.

Prostate cancer is among the most common types of cancer in men. Early detection and proper medical intervention is crucial to ensuring successful treatment. Here we describe a patient clinically presenting with castrate-resistant prostate carcinoma. Comprehensive genomic profiling identified a PTEN inactivating mutation in the patient’s tumor. After being heavily pretreated, the patient showed stable disease on everolimus, a PI3K-Akt-mTOR pathway inhibitor.

Background: The translocation t(15:19) produces the oncogenic BRD4-NUT fusion which is pathognomonic for NUT carcinoma (NC), which is a rare, but extremely aggressive solid tumor. Comprehensive genomic profiling (CGP) by hybrid-capture based next generation sequencing of 186+ genes of a cohort of advanced cancer cases with a variety of initial diagnoses harboring BRD4-NUT may shed further insight into the biology of these tumors and possible options for targeted treatment. Case presentation: Thirty-one solid tumor cases harboring a BRD4-NUT translocation are described, with only 16% initially diagnosed as NC and the remainder carrying other diagnoses, most commonly NSCLCNOS (22%) and lung squamous cell carcinoma (NSCLC-SCC) (16%). The cohort was all microsatellite stable and harbored a low Tumor Mutational Burden (TMB, mean 1.7 mut/mb, range 0–4). In two index cases, patients treated with immune checkpoint inhibitors (ICPI) had unexpected partial or better responses of varying duration. Notably, four cases – including the two index cases - were negative for PD-L1 expression. Neo-antigen prediction for BRD4-NUT and then affinity modeling of the peptide-MHC (pMHC) complex for an assessable index case predicted very high affinity binding, both on a ranked (99.9%) and absolute (33 nM) basis. Conclusions: CGP identifies BRD4-NUT fusions in advanced solid tumors which carry a broad range of initial diagnoses and which should be re-diagnosed as NC per guidelines. A hypothesized mechanism underlying responses to ICPI in the low TMB, PD-L1 negative index cases is the predicted high affinity of the BRD4-NUT fusion peptide to MHC complexes. Further study of pMHC affinity and response to immune checkpoint inhibitors in patients with NC harboring BRD4-NUT is needed to validate this therapeutic hypothesis.

Fusions of the gene RET have been described in thyroid and lung cancers. Here, the AUs identify RET gene alterations, including known fusions, novel fusions, and rearrangements in breast cancer (BC) that are involved in the tumorigenic process and show the benefit of RET therapy in a recurrent BC patient carrying the NCOA4-RET fusion.

Abstract Background This retrospective study was undertaken to determine if the plasma circulating tumor DNA (ctDNA) level and tumor biological features in patients with advanced solid tumors affected the detection of genomic alterations (GAs) by a plasma ctDNA assay. Method Cell-free DNA (cfDNA) extracted from frozen plasma (N = 35) or fresh whole blood (N = 90) samples were subjected to a 62-gene hybrid capture-based next-generation sequencing assay FoundationACT. Concordance was analyzed for 51 matched FoundationACT and FoundationOne (tissue) cases. The maximum somatic allele frequency (MSAF) was used to estimate the amount of tumor fraction of cfDNA in each sample. The detection of GAs was correlated with the amount of cfDNA, MSAF, total tumor anatomic burden (dimensional sum), and total tumor metabolic burden (SUVmax sum) of the largest ten tumor lesions on PET/CT scans. Results FoundationACT detected GAs in 69 of 81 (85%) cases with MSAF > 0. Forty-two of 51 (82%) cases had ≥ 1 concordance GAs matched with FoundationOne, and 22 (52%) matched to the National Comprehensive Cancer Network (NCCN)-recommended molecular targets. FoundationACT also detected 8 unique molecular targets, which changed the therapy in 7 (88%) patients who did not have tumor rebiopsy or sufficient tumor DNA for genomic profiling assay. In all samples (N = 81), GAs were detected in plasma cfDNA from cancer patients with high MSAF quantity (P = 0.0006) or high tumor metabolic burden (P = 0.0006) regardless of cfDNA quantity (P = 0.2362). Conclusion This study supports the utility of using plasma-based genomic assays in cancer patients with high plasma MSAF level or high tumor metabolic burden.

Abstract Background High tumor mutational burden (TMB) is an emerging biomarker of sensitivity to immune checkpoint inhibitors and has been shown to be more significantly associated with response to PD-1 and PD-L1 blockade immunotherapy than PD-1 or PD-L1 expression, as measured by immunohistochemistry (IHC). The distribution of TMB and the subset of patients with high TMB has not been well characterized in the majority of cancer types. Methods In this study, we compare TMB measured by a targeted comprehensive genomic profiling (CGP) assay to TMB measured by exome sequencing and simulate the expected variance in TMB when sequencing less than the whole exome. We then describe the distribution of TMB across a diverse cohort of 100,000 cancer cases and test for association between somatic alterations and TMB in over 100 tumor types. Results We demonstrate that measurements of TMB from comprehensive genomic profiling are strongly reflective of measurements from whole exome sequencing and model that below 0.5 Mb the variance in measurement increases significantly. We find that a subset of patients exhibits high TMB across almost all types of cancer, including many rare tumor types, and characterize the relationship between high TMB and microsatellite instability status. We find that TMB increases significantly with age, showing a 2.4-fold difference between age 10 and age 90 years. Finally, we investigate the molecular basis of TMB and identify genes and mutations associated with TMB level. We identify a cluster of somatic mutations in the promoter of the gene PMS2, which occur in 10% of skin cancers and are highly associated with increased TMB. Conclusions These results show that a CGP assay targeting ~1.1 Mb of coding genome can accurately assess TMB compared with sequencing the whole exome. Using this method, we find that many disease types have a substantial portion of patients with high TMB who might benefit from immunotherapy. Finally, we identify novel, recurrent promoter mutations in PMS2, which may be another example of regulatory mutations contributing to tumorigenesis.

After failure of anthracycline- and platinum-based therapy, no effective therapies exist for management of metastatic triple-negative breast cancer (TNBC). We report a case of metastatic TNBC harboring MCL1 amplification, as identified by comprehensive genomic profiling in the course of clinical care. MCL1 is an antiapoptotic gene in the BCL2 family, and MCL1 amplification is common in TNBC (at least 20%). A personalized dose-reduced regimen centered on a combination of sorafenib and vorinostat was implemented, based on preclinical evidence demonstrating treatment synergy in the setting of MCL1 amplification. Although hospice care was being considered before treatment initiation, the personalized regimen yielded 6 additional months of life for this patient. Further rigorous studies are needed to confirm that this regimen or derivatives thereof can benefit the MCL1-amplified subset of TNBC patients.

Cervical cancer epitomizes the success of cancer prevention through the human papillomavirus (HPV) vaccine, but significant challenges remain in the treatment of advanced disease. We report the first three cases of cervical carcinoma harboring an FGFR3–TACC3 fusion, which serves as a novel therapeutic target. The fusion, identified by comprehensive genomic profiling, activates the FGFR pathway that has been implicated in HPV-driven carcinogenesis. One of the patients whose tumor contained the FGFR3–TACC3 fusion was treated with an investigational FGFR tyrosine kinase inhibitor. Concomitant molecular alterations involving the PI3K/AKT/mTOR and RAF/MEK pathways were also identified and suggest other treatment strategies that deserve investigation. This case series highlights the role of comprehensive genomic profiling in the identification of new therapeutic targets and in targeted therapy selection for patients with cervical cancer.

Suspected metastatic site lesions that are poorly differentiated present a diagnostic challenge when morphologic and immunohistochemical profiling cannot establish the primary tumor site. Here we present a patient diagnosed with both a malignant neoplasm in the lung and a right upper extremity (RUE) neoplasm of unclear histogenetic origin. Immunohistochemical staining performed on the latter specimen was inconclusive in determining the site of origin. Although the lung biopsy sample was insufficient for molecular testing, hybrid capture-based comprehensive genomic profiling (FoundationOne) identified an EML4-ALK rearrangement in the RUE lesion. Crizotinib treatment resulted in a major response in both the RUE and the lung lesions. This report illustrates the utility of comprehensive genomic profiling employed at the initial presentation of an unknown primary malignant neoplasm, which resulted in the front-line use of targeted therapy and a significant and sustained antitumor response.

Background: Carcinoma of unknown primary (CUP) accounts for 3-5% of all adult solid tumors. An extensive search for the anatomic site of origin is often undertaken in an attempt to tailor systemic treatment, but the latter often has limited efficacy - especially in the setting of an initial treatment failure. Molecularly targeted therapy is an emerging approach that may offer greater efficacy and less toxicity but is most likely to be effective when pairing a tumor harboring a sensitizing genomic alteration with an agent directed at the altered gene product. We report a patient with a CUP harboring a MET amplification with a complete metabolic response to crizotinib despite also harboring a KRAS mutation. Methods: Genomic profiling was performed using a clinical next-generation-sequencing-based assay, FoundationOne®, in a CAP-accredited laboratory certified by Clinical Laboratory Improvement Amendments (Foundation Medicine, Cambridge, Mass., USA). Results: The CUP harbored both MET amplification (16 copies) and a KRAS G12V mutation. The patient was treated with crizotinib, a MET inhibitor, and has experienced a complete normalization of tumor metabolic activity for more than 19 months. Conclusions: Genomic profiling of CUP may reveal clinically meaningful genomic alterations that can guide targeted therapy decision-making. The use of this approach should be studied prospectively as a strategy for the effective treatment of CUP patients and for avoiding resource-intensive workups to identify the tumor site of origin.

Context: For patients with metastatic papillary thyroid carcinoma (PTC) refractory to radioactive iodine (RAI) treatment, systemic chemotherapy has limited efficacy. Such tumors frequently harbor BRAF V600E, and this alteration may predict responsiveness to vemura-fenib treatment. Objective: We report a metastatic PTC patient refractory to RAI treatment that underwent genomic profiling by next-generation sequencing. The sole genomic alteration identified was BRAF V600E on a near diploid genome with trisomy 1q. With vemurafenib treatment, the patient experienced a dramatic radiographic and clinical improvement, with the duration of an ongoing antitumor response exceeding 23 months. Design: Hybridization capture of 3,769 exons of 236 cancer-related genes and the introns of 19 genes frequently rearranged in cancer was applied to >50 ng of DNA extracted from a formalin-fixed, paraffin-embedded biopsy of a lymph node containing metastatic PTC and was sequenced to a high, uniform coverage of ×616. Results: A BRAF V600E alteration was identified with no other somatic genomic alterations present within a near diploid tumor genome. The patient initially received vemurafenib at 960 mg twice daily that was reduced to 480 mg twice daily due to rash and diarrhea and has experienced an ongoing antitumor response exceeding 23 months by both PET-CT and dedicated CT imaging. Conclusions: Genomic profiling in metastatic, RAI-refractory PTC can reveal a targetable BRAF V600E alteration without compounding somatic alterations, and such patients may derive a more prolonged benefit from vemurafenib treatment. Prospective clinical trials are ongoing to confirm our preliminary observation.