Thoracic Radiotherapy and Inhibition of PD-1 and LAG-3 for Locally Advanced Non-Small Cell Lung Cancer

Non-small cell lung cancer (NSCLC) represents more than 80% of lung cancers, and approximately 35% of NSCLC patients present with stage III disease. Standard treatment for patients with locally advanced NSCLC, which may be defined as American Joint Committee on Cancer (AJCC) stage III disease or unresectable stage II disease, typically consists of conventionally fractionated (1.8-2.0 Gray (Gy) per day) radiotherapy (RT) to a total dose of approximately 60 Gy with concurrent chemotherapy. Based on the PACIFIC trial (NCT02125461), patients without disease progression after concurrent chemoradiotherapy are typically offered a one-year course of the adjuvant PD-L1 inhibitor durvalumab, regardless of PD-L1 tumor proportion score (TPS). Recent results from the PACIFIC-2 trial (NCT03519971) failed to demonstrate a survival benefit of adding concurrent durvalumab to chemoradiotherapy, and therefore the PACIFIC regimen of adjuvant immunotherapy following chemoradiation remains the standard of care for unresectable LA-NSCLC. Based on the PACIFIC trial standard, both the RT and systemic therapy utilized to treat LA-NSCLC patients in standard of care follow a non-biomarker selected "one-size-fits-all" approach in the US. In advanced NSCLC, however, patients with high PD-L1 TPS score (≥ 50%) benefit more from the immune checkpoint inhibitors pembrolizumab, atezolizumab, and cemiplimab compared to cytotoxic chemotherapy alone, and single agent immunotherapy (IO) has become a first-line standard of care for such patients. Omitting chemotherapy from the treatment of LA-NSCLC patients with high PD-L1 expression who will receive IO is a logical step. On the other hand, patients with low PD-L1 TPS score (\< 50%) benefit less from single agent IO in advanced NSCLC, and IO is typically offered together with chemotherapy. Given the poor response to IO for patients with low PD-L1 TPS expression, the investigator team believes that combination IO plus chemotherapy could be particularly beneficial in the low-TPS LA-NSCLC setting. Induction (also referred to as neoadjuvant) IO before RT for LA-NSCLC has several potential advantages. For example, the native tumor functioning as an in-situ vaccine in the neoadjuvant setting, with an intact lymphatic system and unirradiated nodal status, could promote optimal immune priming. Indeed, neoadjuvant IO has shown impressive efficacy in both deficient mismatch repair (dMMR) rectal cancer and melanoma. For early-stage NSCLC, this approach is highlighted in recent neoadjuvant trials of immune checkpoint inhibition in combination with chemotherapy for resectable NSCLC, such as CheckMate 816 (NCT02998528), Keynote-671 (NCT03425643), and NADIMII (NCT03838159). Furthermore, results from the PACIFIC trial and real-world data show that approximately 25% of LA-NSCLC patients started on chemoradiotherapy never receive consolidation durvalumab, and approximately 50% of patients started on consolidation durvalumab do not complete the intended year of treatment. Thus, administering induction IO may be particularly advantageous for LA-NSCLC patients to enhance IO treatment delivery. SPRINT (NCT03523702) was a multi-institutional Phase II trial where biomarker-selected (PD-L1 TPS ≥ 50%) LA-NSCLC patients were treated with 3 cycles of induction pembrolizumab to reduce the extent of thoracic RT, and cytotoxic chemotherapy was omitted from the treatment regimen. Patients with PD-L1 TPS \< 50% received standard concurrent chemoradiotherapy (chemoRT) followed by standard adjuvant treatment, to serve as a non-randomized comparator for evaluating toxicities and clinical outcomes. Following induction, PD-L1 TPS ≥ 50% patients received PET-based dose-painted RT according to a novel and personalized approach that was studied in two prior trials (NCT02073968, NCT03481114) that can decrease toxicities by utilizing shortened courses of RT and reducing doses administered to small tumors and nodes. Based on these studies, PET response to induction IO using PERCIST criteria may serve as a useful prognostic factor to identify LA-NSCLC patients more likely to respond to novel IO regimens at an earlier time point in treatment (prior to administration of RT). This observation is also supported by prior prospective and retrospective studies showing that response on PET rather than CT may be a better predictor of survival for NSCLC patients receiving IO. Furthermore, early identification of IO response provides added insight to a tumor's unique immune biology that could be used to individualize subsequent RT, IO, and/or chemotherapy treatment plans. Overall, SPRINT demonstrated high treatment efficacy with limited toxicity and introduces PET response to induction IO as a novel endpoint that allows for early assessment of treatment activity, and which could potentially serve as a new paradigm in LA-NSCLC interventional studies. Based on this experience, the investigator team believes that the SPRINT treatment approach (induction IO followed by RT and consolidation IO) can be used to investigate other novel IO combinations for biomarker-selected LA-NSCLC. One such immune target is lymphocyte activation gene 3 (LAG3), which is expressed by various immune cells, and regulates effector T-cell activation and responses. LAG-3 inhibition restores the effector function of exhausted T cells, enhancing their ability to attack tumor cells. In addition, LAG-3's inhibitory effects on T cells appear distinct from those of PD-L1, serving as a potential rationale for combining LAG-3 and PD-1/PD-L1 inhibition. In metastatic melanoma, the RELATIVITY-047 trial (NCT03470922) showed that combination PD-1 and LAG-3 blockade with nivolumab and relatlimab, respectively, improves progression-free survival (PFS) compared to nivolumab alone with a favorable toxicity versus nivolumab plus the CTLA-4 inhibitor ipilimumab. These findings suggest that combination PD-1 and LAG-3 inhibition may have potential utility for treatment of other malignancies, such as NSCLC. Furthermore, a Phase I study of cemiplimab and fianlimab in unresectable stage IIIB-C or IV NSCLC (NCT03005782) showed that the combination demonstrated clinical activity with a similar safety profile compared to cemiplimab alone. Given the promising findings observed in SPRINT, RELATIVITY-047, and NCT03005782, the research team believes that dual immune checkpoint blockade with the PD-1 inhibitor cemiplimab plus the LAG-3 inhibitor fianlimab administered before and after thoracic RT, as per the SPRINT approach, has valuable therapeutic potential for LA-NSCLC patients. Treatment will be administered in three phases as detailed in this registration: (1) three cycles of induction IO, with histology-specific platinum doublet chemotherapy (PDC) added for subjects with PD-L1 TPS \<50%; (2) thoracic RT (with concurrent PDC recommended for subjects with PD-L1 TPS \<50%); (3) 13 cycles of consolidation IO.