The successes of immunotherapy and targeted therapy in metastatic non-small cell lung cancer (NSCLC) have raised hopes for implementing these therapies in earlier stages of the disease. Recent trial results indicate that neoadjuvant, adjuvant, and perioperative immunotherapy may improve the cure rates of early-stage NSCLC. While early diagnosis of patients could make immunotherapy a significant option for early-stage NSCLC, identifying reliable biomarkers remains a challenge.

Key Points of the Article

• Immunotherapy and targeted therapy show promising prospects for treating early-stage resectable NSCLC.
• Identifying early-stage patients who can benefit from aggressive treatment is challenging, hindering the implementation of these new therapies.
• Pathologic response and ctDNA are potential biomarkers for allocating appropriate treatment strategies, but more data is needed to support personalized treatment and maximize the chances of cure.

Identifying Early-Stage Lung Cancer Patients

A critical issue in treating early-stage NSCLC is identifying patients at a resectable stage. Currently, over 40% of patients are diagnosed with metastatic disease. The guidelines from the U.S. Preventive Services Task Force recommend screening only for patients with a smoking history of more than 20 pack-years and/or those who quit smoking within the past 15 years. These guidelines may not cover patients who have smoked for a long time but less than 20 pack-years, potentially exacerbating gaps in lung cancer screening.

A study presented at the 2024 ASCO Annual Meeting evaluated the effectiveness of expanding lung cancer screening criteria. The researchers increased the detection rate of early-stage lung cancer and reduced disparities among female and Black smoking patients. Similarly, patients with targetable mutations in NSCLC tend to be younger and smoke less, making them difficult to capture using current screening guidelines. Studies like TALENT and FANSS have the potential to help expand lung cancer screening criteria to include patients more likely to develop targetable mutations, thereby promoting better adjuvant treatment for cancers driven by oncogenic mutations. Better identification of early-stage disease can increase the proportion of patients receiving curative treatment and improve survival rates.

Adjuvant Therapy vs. Neoadjuvant Therapy vs. Perioperative Strategies

For early-stage, resectable NSCLC, various treatment strategies—whether adjuvant, neoadjuvant, or perioperative—have the potential to improve relapse-free survival and overall survival (OS). Adjuvant immunotherapy trials, such as PEARLS/KEYNOTE-091 and IMpower010, have shown disease-free survival (DFS) benefits, and the ADAURA trial demonstrated that adjuvant osimertinib targeted therapy can bring DFS and OS benefits.

Subsequently, trials using neoadjuvant strategies to treat early-stage NSCLC have shown improvements in DFS. Based on data from neoadjuvant and adjuvant therapy trials, research into perioperative treatment strategies—treating patients before and after surgical resection—has begun. So far, only one perioperative immunotherapy trial (KEYNOTE-671) has demonstrated an OS benefit. Other trials (AEGEAN, CheckMate-77T, and Neotorch) have indeed shown that perioperative immunotherapy improves DFS compared to chemotherapy alone, but OS data remains immature. The surge in perioperative research raises the question: how can we identify patients who benefit most from the adjuvant component of perioperative treatment?

Biomarkers and Future Directions

Pathologic response is a promising biomarker for identifying patients who may not need adjuvant therapy. Based on the proportion of residual viable tumor cells (%RVT), patients can be classified as achieving pathologic complete response (pCR; %RVT=0), major pathologic response (MPR; %RVT<10%), or pathologic non-response (%RVT>10%). In neoadjuvant trials, patients who achieved pCR or MPR had better outcomes, regardless of treatment group. This finding suggests that pathologic response as a biomarker could serve as a surrogate endpoint for OS, allowing clinical trials to obtain results more quickly. A meta-analysis of existing trial data indicates that at the patient level, pCR and MPR are strong surrogates for event-free survival (EFS); however, at the trial level, EFS and OS surrogacy is only moderate, partly due to trial heterogeneity and immature data.

Another potential biomarker that could be used to “identify patients needing intensified treatment” is ctDNA. Patients with Stage I NSCLC in whom ctDNA is detectable before surgery are more likely to experience recurrence. For patients receiving neoadjuvant therapy, pre-surgery ctDNA clearance is associated with pathologic response and DFS improvement. Therefore, “monitoring ctDNA dynamics” could be used for patient risk stratification, although trials are needed to validate this hypothesis.

However, some potential confounding factors could delay ctDNA monitoring of treatment response. First, the limit of detection may not be sufficient. Data from other diseases, particularly multiple myeloma, indicate that sufficient depth of detection is necessary to adequately monitor residual lesions. Second, genotypic differences may be associated with higher ctDNA detection rates, but it is unclear whether this is due to more proliferative tumors (where necrotic cells release DNA), more aggressive tumors (with more circulating tumor cells), or characteristics of genomic instability (leading to increased DNA release).

Currently, there is a lack of reliable data to make treatment decisions based on pathologic response and ctDNA clearance, particularly at the population level. The observed issue when weighing whether neoadjuvant or perioperative therapy is more suitable for early-stage NSCLC is that, despite significant DFS improvements, OS data remains largely immature, and no trials have directly compared neoadjuvant and perioperative treatment strategies. Therefore, clinical trials are needed where patients receive neoadjuvant therapy and are then randomized to adjuvant therapy or placebo, with stratification based on pathologic response and/or ctDNA clearance rates. This would better demonstrate the predictive value of these endpoints (pathologic response and ctDNA clearance rates as moderate survival surrogates at the trial level). Once this data is available, we can determine the best way to personalize treatment to maximize the chances of curing NSCLC patients.

Robert B. Cameron

Department of Medicine, Section of Hematology/Oncology, University of Chicago