Yi, Eunhee

Abstract Objectives Sclerosing pneumocytomas are rare pulmonary neoplasms that are typically benign. However, rare patients experience progressive disease, and therapy targeting specific genetic underpinnings could be an attractive therapeutic option. Recent studies have found recurrent AKT 1 mutations in sclerosing pneumocytoma, but little is known about whether oncogenic fusion genes may also be present. Methods To better understand the genetic background, 10 sclerosing pneumocytomas were subjected to next-generation sequencing cancer mutation panel testing (n = 9) and/or RNA sequencing (n = 3). The patients were all women (average age, 47 years; range, 17-74 years). Results Eight patients had solitary sclerosing pneumocytomas, while one had two tumors, and one had many bilateral tumors. Recurrent mutations were noted in genes involved in the mTOR pathway, including AKT1, PIK3R1, and PTEN. AKT1 alterations were particularly common, present in 78%. No recurrent genetic fusions were identified. The patient in our study with multiple bilateral lesions was treated with the mammalian target of rapamycin (mTOR) inhibitor everolimus, with no objective radiographic evidence of treatment response after 4 months. Conclusions Our data further support that abnormal activation of the mTOR pathway is a consistent genetic event in sclerosing pneumocytoma. This warrants further exploration to determine if mTOR pathway inhibitors may be effective in patients with metastatic or recurrent disease.

3122 Background: The current standard work-up for both diagnosis and predictive biomarker testing in metastatic non-small cell lung cancer (NSCLC), can exhaust an entire tumor specimen. Notably, gene mutation panels or tumor mutation burden (TMB) testing currently requires 10 tissue slides and ranges from 10 days to 3 weeks from sample acquisition to test result. As more companion diagnostic (CDx)-restricted drugs are developed for NSCLC, rapid, tissue-sparing tests are sorely needed. We investigated whether TMB, T-effector (TEFF) gene signatures and PD-L1 status can be inferred from H&E images alone using a machine learning approach. Methods: Algorithm development included two steps: First, a neural network was trained to segment hand-annotated, pathologist-confirmed biological features from H&E images, such as tumor architecture and cell types. Second, these feature maps were fed into a classification model to predict the biomarker status. Ground truth biomarker status of the H&E-associated tumor samples came from whole exome sequencing (WES) for TMB, RNAseq for the TEFF gene signatures or reverse-phase protein array for PD-L1. Digital H&E images of NSCLC adenocarcinoma for model development were obtained from the cancer genome atlas (TCGA) and commercial sources. Results: This approach achieves > 75% accuracy in predicting TMB, TEFF and PD-L1 status, offers a way to interpret the model, and provides biological insights into the tumor-host microenvironment. Conclusions: These findings suggest that biomarker inference from H&E images is feasible, and may be sufficiently accurate to supplement or replace current tissue-based tests in a clinical setting. Our approach utilizes biological features for inference, and is thus robust, interpretable, and readily verifiable by pathologists. Finally, biomarker status inference from a single H&E image may enable testing in patients whose tumor tissue has been exhausted, spare further tissue use, and return test results within hours to enable rapid treatment decision-making to maximize patient benefit.