ABSTRACT Lung cancer is the leading cause of cancer-related mortality in the United States and worldwide. The efficacy of immune checkpoint inhibitors (ICIs) in patients with metastatic non-small lung cancer (NSCLC) prompted the clinical investigation of these agents in the early-stage operable setting. Several theoretical advantages exist when we administer ICIs before surgery (neoadjuvant) rather than postoperatively (adjuvant), including an opportunity to address micrometastases early in the course of treatment, and may impart immunologic memory to prevent tumor recurrence. Indeed, the results from our preclinical models of resectable NSCLC demonstrated that combined neoadjuvant ICIs resulted in fewer lung metastases, greater immune infiltration of tumors, and longer overall survival compared with mice treated with monotherapy or adjuvant combined ICIs. Those results informed the first reported randomized phase 2 study testing neoadjuvant ICI combinations in patients with resectable NSCLC using major pathologic response (MPR, ≤10% viable tumor) as a surrogate endpoint for clinical efficacy (NEOSTAR, PI: Cascone). Neoadjuvant chemoimmunotherapy has been shown to be highly promising for resectable NSCLC, and is now being tested in one of the phase 3 randomized studies in patients with operable NSCLC (CheckMate-77T, Lead PI: Cascone). However, a major shortcoming of all of the neoadjuvant trials, is that no validated biomarker exists that can be used to stratify patients. Consequently, many of these patients on these trials do not achieve an MPR at surgery, indicating that limited benefit may be gained from induction ICIs. By delaying surgery in patients who may not benefit, the risks of disease progression and of eliminating a chance to offer potentially curative surgery upfront occur. The ongoing evaluation of molecular biomarkers of clinical benefit to ICIs has proved disappointing as evidenced by the significant intertrial variability, possibly related to intratumor heterogeneity. By contrast, radiologic imaging provides a holistic view of tumor characteristics and interactions with the adjacent tissue. Built on our promising preliminary data, we propose to spearhead radiographic and radiogenomics strategies to address this unmet clinical need. We hypothesize that imaging phenotypes reflect tumor microenvironment, and quantitative imaging phenotyping will shed light on our understanding of the mechanisms of response to ICIs and yield surrogates of clinical efficacy. We will leverage the parallel assessment of well-curated data from unique clinical trials and immunocompetent mouse models to develop new imaging biomarkers and validate their clinical and biological relevance. The strength of this proposal is our interdisciplinary team with the requisite expertise and ability to treat patients, obtain and analyze high- quality, longitudinal imaging and biospecimens and rapidly evaluate putative imaging biomarkers for therapeutic response and clinical outcomes. The advent of imaging biomarkers will: 1) identify those patients most likely to benefit from neoadjuvant ICIs, 2) maximize the clinical effectiveness, and 3) lead to the development of new therapies that will improve outcomes for a greater number of patients with resectable NSCLC.

Lay Summary This project aims to spearhead the role of imaging in the personalized management of early-stage lung cancer treated with neoadjuvant immunotherapy. The proposed studies will lead to the development of computational tools to extract quantitative imaging biomarkers from serial PET and CT scans and to the evaluation of the biological and clinical relevance of these imaging biomarkers. The new imaging biomarkers will close the unmet clinical need in immunotherapy for early-stage lung cancer by identifying those patients most likely to benefit from treatment, maximizing the clinical effectiveness of immunotherapies, and thus promoting a major paradigm shift of the role of radiographic imaging to serve as an important approach in precision cancer management.