Neoadjuvant Versus Adjuvant Systemic Therapy for Early-Stage Non-Small Cell Lung Cancer: The Changing Landscape Due to Immunotherapy

Abstract

Non-small cell lung cancer (NSCLC) remains a major cause of morbidity and mortality worldwide. One-third of NSCLC patients present with surgically resectable, non-metastatic disease; however, many of these patients will recur despite curative surgery and adjuvant therapy. The recent publication of randomized trials incorporating immune check-point inhibitors (ICI) to the standard neo-adjuvant and adjuvant treatment regimens has reported improved survival with manageable toxicity profiles. The IMpower 010 studied the use of adjuvant atezolizumab after standard surgery and adjuvant chemotherapy. They demonstrated an improvement in 3-year disease-free survival (DFS) prompting a change in treatment guidelines. The Checkmate 816 and NADIM II studies evaluated the addition of pembrolizumab and nivolumab, respectively, to standard neo-adjuvant chemotherapy. The results from both trials showed an improvement in 2-year event-free survival (EFS) and 2-year PFS (PFS), respectively. In this review, we summarize the prior data regarding adjuvant and neo-adjuvant chemotherapy in NSCLC and elaborate on results from the newer trials incorporating ICIs. We briefly discuss the pros and cons of each treatment approach along with areas that need further clarity to inform clinical practice and future directions for research in this disease.

This review summarizes prior data regarding adjuvant and neo-adjuvant chemotherapy in non-small cell lung cancer and elaborates on results from the newer trials incorporating immune checkpoint inhibitors. The pros and cons of each treatment approach are briefly discussed, including areas that need further clarity to inform clinical practice and future directions for research in this disease.

Implications for Practice.

The incorporation of immune checkpoint inhibitors (ICIs) as part of adjuvant or neo-adjuvant therapy for operable non-small cell lung cancer (NSCLC) should be strongly considered in light of accumulating evidence of benefit. Although pre-clinical data and evidence from other cancers suggest an advantage of using ICIs in the neo-adjuvant setting, clinical evidence with respect to overall survival in NSCLC is not mature. Surrogate biological markers, such as pathological complete response and major pathological response, are being used in neo-adjuvant trials with promising correlations with disease-free survival. The use of biomarkers such as ctDNA may allow for more biomarker selected personalized treatment approaches in the future.

Introduction

Lung cancer remains a major cause of cancer-related death worldwide.¹ The majority of lung cancers are non-small cell lung cancers (NSCLC), of which one-third present with surgically resectable disease,² with gradual improvements in survival trends over time.³ However, despite complete tumor resection, up to 60% of patients will recur, most often with distant metastasis and resultant poor overall survival (OS),⁴ underscoring the need for additional therapy beyond surgery. The rationale for perioperative systemic therapy is to address micrometastasis to prevent distant metastasis. This additional systemic therapy can be delivered either before surgery (neo-adjuvant) or after surgery (adjuvant). Adjuvant chemotherapy is the current standard of care for surgically resected stage IB-IIIA NSCLC. The use of neoadjuvant chemotherapy previously yielded modest benefits and was not recommended as standard of care, even though there were no tangible differences in OS between neoadjuvant and adjuvant chemotherapy approaches.^5,6 Recent improvements in our understanding of cancer biology, particularly the role of the immune system and the availability of immune checkpoint inhibitors (ICI) have allowed the field to revisit neoadjuvant therapy. ICIs result in durable responses and longer survival in 20%-30% of patients with advanced NSCLC⁷ leading to interest in the incorporation of these treatments in early-stage NSCLC. Results from neoadjuvant studies incorporating ICIs reveal impressive pathological response and progression-free survival (PFS). This review will focus on the role of neoadjuvant and adjuvant ICIs either alone or in combination with chemotherapy in the treatment of resectable NSCLC. We will compare these newer approaches with existing standard of care recommendations and discuss future directions of the treatment landscape in this space.

Adjuvant Therapy in NSCLC

Adjuvant Chemotherapy Trials

Several phase II adjuvant chemotherapy trials in early 90s showed modest benefit. A meta-analysis of 52 randomized controlled trials (RCTs) comprising 9387 patients revealed that surgery and chemotherapy had a risk ratio of 0.87 (0.74-1.02, P = .08) as compared to surgery alone, with the caveat that this benefit only extended to cisplatin containing regimens. However, they failed to show a statistically significant improvement in 5-year OS.⁸ The first major evidence of benefit with the use of adjuvant chemotherapy in NSCLC came from the randomized International Adjuvant Lung Cancer Trial (IALT) that demonstrated a statistically significant improvement in OS with adjuvant cisplatin-based chemotherapy in surgically resected stages I–III NSCLC, with an absolute survival benefit of 4.1% at 5 years.⁹ The JBR10 trial randomized 482 patients with resected stage IB/II NSCLC to adjuvant chemotherapy with cisplatin and vinorelbine or observation and demonstrated a 15% improvement in 5-year survival.¹⁰ The ANITA trial, also showed a benefit of 6% at 5 years with adjuvant cisplatin and vinorelbine among stages IB-III.¹¹ These positive RCTs formed the backbone for the positive results noted in the LACE meta-analysis, which demonstrated an absolute survival benefit of 5.4% at 5 years (6.9% decrease in lung cancer-related death with a 1.4% increase in non-lung cancer-related death).¹² These findings established the role of adjuvant chemotherapy as the standard of care in stages IB-III NSCLC. Incorporation of newer chemotherapy agents, different doses, or schedules failed to improve this 4%-5% survival benefit.^13,14 It was clear that newer strategies were required to improve on this real, yet modest benefit from adjuvant chemotherapy.

Adjuvant Targeted Therapy in NSCLC

The first oncogenic driver discovered in NSCLC involved the epidermal growth factor receptor (EGFR), activated by mutations in the ATP binding domain of the EGFR kinase. Gefitinib was the first tyrosine kinase inhibitor (TKI) that reversibly inhibits the ATP binding site. ADJUVANT-CTONG 1104 and IMPACT (WJOG6410L) evaluated the role of adjuvant gefitinib in patients with completely resected NSCLC with sensitizing EGFR mutations. Patients were randomized to receive adjuvant gefitinib for 2 years or 4 cycles of cisplatin and vinorelbine. Although both trials demonstrated an initial improvement in PFS, neither trial demonstrated a sustained OS benefit at 5 years.^15,16 Better results emerged with the third generation EGFR TKI osimertinib, which targets the cysteine-797 residue in the ATP binding site via irreversible covalent bond formation, thereby inhibiting the sensitizing EGFR mutations as well as the resistant T790M mutation.¹⁷ The ADAURA phase III trial randomized patients with completely resected stage II and III NSCLC with sensitizing EGFR mutations to adjuvant osimertinib or placebo for 3 years. The trial demonstrated improved DFS with 98% of the patients in the osimertinib group (95% CI, 95-99) vs. 85% of those in the placebo group (95% CI, 80-89) alive without CNS disease at 2 years (overall hazard ratio 0.18; 95% CI, 0.10-0.33). The OS results for this trial are awaited.¹⁸ However, updated evidence (IASCLC 2022) with adjuvant osimertanib points to a continued improvement of DFS at 5 years and most importantly a decrease in the rate of brain and spine metastasis, particularly in stages II-IIIA resected NSCLC. Data for the use of other targeted agents in the adjuvant treatment setting are awaited. ALCHEMIST (Adjuvant Lung Cancer Enrichment Marker Identification and Sequencing Trial) is a clinical trial for patients with completely resected stages IB-IIIA NSCLC. One arm compares the efficacy of adjuvant crizotinib for 2 years versus placebo for NSCLC bearing ALK rearrangements.¹⁹ These data suggest that patients with NSCLCs bearing oncogenic drivers may benefit from adjuvant therapy with targeted therapies However, longer follow-up and data on OS are necessary before widespread adoption as standard of care.

Adjuvant Immunotherapy/Chemo-Immunotherapy in Early NSCLC

The IMpower010 trial randomized 1005 patients with completely resected stages IB-IIIA NSCLC after 4 cycles of cisplatin-based chemotherapy to atezolizumab for 1 year or best supportive care (BSC). Patients were eligible for inclusion irrespective of EGFR/ALK mutation status or PDL-1 expression. The primary end-point of the study was DFS, evaluated hierarchically in the stages II-IIIA population with PDL1 expression > 1%, followed by all stages II-IIIA patients, and lastly in the intention to treat (ITT) population. This study achieved its primary endpoint by demonstrating a statistically significant improvement in 3-year DFS in the PDL-1 >1% population from 48% to 60% (stratified HR 0·66.95% CI, 0.50-0.88; P = .0039). The 3-year DFS in the entire stages II-IIIA population improved from 49% to 56%, stratified HR 0·79 (0.64-0.96; P= 0.020). The subgroup analysis demonstrated a graded increase in benefit with increasing PDL-1 expression, with no difference in outcomes detected based on the presence/absence of EGFR or ALK mutations.²⁰ This study led to FDA approval of adjuvant atezolizumab for 1year following surgical resection of stages IB-IIIA NSCLC.²¹ The DFS benefit seen in this study was primarily driven by the PDL-1 >50% cohort as demonstrated in a post hoc exploratory subgroup analysis (unstratified HR 0.43; 95% CI, 0.26-0.71). Based on this, the European Commission approved atezolizumab for use as an adjuvant therapy following complete resection and platinum-based chemotherapy in adults with NSCLC whose tumors do not have EGFR mutations or ALK alterations, but have a PDL-1 expression of 50% or higher.²² Although the OS data remain immature for this population, a recent update demonstrated a trend toward OS improvement with atezolizumab. At 5 years, the OS rates were 76.8% and 67.5% in the atezolizumab arm and the BSC arm respectively, with median OS not reached in either arm.²³

The phase III PEARLS/Keynote 091 trial randomized 1177 patients with stage IB-IIIA completely resected NSCLC after adjuvant chemotherapy to pembrolizumab versus placebo for 18 cycles. This study had dual primary endpoints of DFS in the ITT population as well as DFS in those with PDL1-TPS > 50%. The study stratified for PDL1 prior to randomization and allowed entry of EGFR and ALK mutated NSCLC. A recently published interim analysis of this trial demonstrated an improvement in median DFS from 42 months to 53.6 months (HR 0·76 [95% CI, 0.63-0.91], P = .0014). In the PDL1-TPS >50% population, the median DFS had not been reached after 35.6 months (IQR 27.1-45.5) of follow-up.²⁴ Two other trials are evaluating the role of checkpoint inhibitors as adjuvant therapy. The ANVIL study, a part of the ALCHEMIST group of trials is an ongoing randomized trial evaluating the role of adjuvant nivolumab up to 1 year in stages IB-IIIA completely resected ALK and EGFR wild-type NSCLC treated with adjuvant chemotherapy.²⁵ The BR31 is a similar ongoing phase III randomized trial evaluating the role of 1 year of adjuvant durvalumab after resection and adjuvant chemotherapy. Patients are eligible for recruitment irrespective of PDL1 expression and EGFR or ALK mutation.²⁶ While the OS data are awaited in above trials, early DFS benefit appears promising.

Taken together, adjuvant therapy with osimertanib is recommended for patients with surgically resected stage IB-IIIA NSCLC, with the greatest magnitude of DFS benefit in stage II-IIIA NSCLC patients with EGFR mutations. In patients without EGFR or ALK mutations, atezolizumab is recommended in the adjuvant setting for resected stage IB-IIIA NSCLC, with the greatest magnitude of DFS benefit in patients whose NSCLCs expressed PDL-1 > 50%. Pembrolizumab could be considered as a reasonable alternative in this population. We will now turn to advances in neoadjuvant systemic therapies for resectable NSCLC. For comparison of systemic therapy between neoadjuvant and adjuvant approaches, we have not included neoadjuvant approaches that incorporate radiation. The individual trial details of adjuvant chemotherapy and immunotherapy or adjuvant dual check-point blockade is presented in Table 1.

Table 1.

Prospective trials of adjuvant chemotherapy + immunotherapy OR dual-check-point blockade.

Trial name, sample size	Design, phase, disease stage	Intervention	Primary outcome	Results	Survival	Comments
Impower01020 N = 1005	Randomized controlled trial, phase III, Stage IB-IIIA	Cisplatin-based chemotherapy × 4 followed by atezolizumab 1200 mg q21 days × 16 versus cisplatin-based chemotherapy × 4 followed by BSC	DFS in stages II-IIIA with PDL1 > 1% DFS in the stages II-IIIA population DFS in the ITT population		Three-year DFS in stages II-IIIA with PDL1 > 1%—60% in atezo arm vs. 48% in BSC arm HR 0·66 (95% CI, 0·50-0·88; P = .0039) Stages II-IIIA—56% in atezo arm and 49% in the BSC arm HR0·79 (0·64-0·96; P = .020) ITT population-58% in atezo arm and 53% in the BSC arm HR 0·81 (0·67-0·99; P = .040)	First reported phase III adjuvant chemo+ICI study. No statistically significant DFS benefit for the ITT population, but beneficial difference was noted in stage II and IIA populations. Improved outcomes in the PDL1 > 1% population
ANVIL NCT0259594425	RCT, Phase III	Nivolumab q 3 weeks up to 1 year versus observation	DFS OS	Awaited	Awaited	EGFR− ALK−
PEARLS/ Keynote 09124 N-1177	RCT, Phase III, Stage IB-IIIA	Pembrolizumab 200 mg q3 weeks up to 1 year versus observation (after adjuvant chemo)	DFS in ITT population DFS in PDL1-TPS > 50%		mDFS in the ITT population- 53.6 vs. 42 (HR 0.76 [95% CI 0·63-0·91], P = .0014) mDFS in the PDL1 > 50%—not reached in either arm	All histology Irrespective of PDL1 EGFR-7% ALK 1%
BR31 NCT0227337526	RCT, Phase III	Durvalumab up to 1 year versus Placebo (after adjuvant chemo)	DFS among those with PDL1 TPS > 25%	Awaited	Awaited	EGFR-neg ALK-neg
MERMAID-1 NCT0438536827 N = 332	RCT, Phase III, Stage II-III NSCLC who have completed curative surgery, with no evidence of disease recurrence and are minimal residual disease (MRD) positive after surgery	Durvalumab + standard of care chemotherapy versus placebo + standard of care chemotherapy	DFS among the MRD positive patients	Awaited	Awaited
MERMAID-2 NCT0464246928	RCT, Phase III Stage II-III NSCLC who have completed curative intent therapy (complete resection ± neoadjuvant and/or adjuvant chemotherapy) with no RECIST v1.1 identified disease and become MRD + during a 96-week surveillance period	Durvalumab versus placebo	DFS in PDL-1 TC > 1%	Awaited	Awaited

Neoadjuvant Therapy in NSCLC

The Biological Rationale and Advantages of Neoadjuvant Therapy

The rationale for use of neoadjuvant therapy is based on theoretical considerations, including, (1) tumor downsizing that can lead to increased R0 resection rates,²⁹ (2) early treatment of micrometastatic disease may lead to improvements of DFS and OS³⁰ and (3) potential to utilize in vitro drug susceptibility testing to inform further therapy for those demonstrating a poor response to initial neoadjuvant therapy.³¹ More recently, the rationale to use neoadjuvant ICI-based regimens includes the idea that the intact tumor and primary lymphatics allow better T-cell priming, with the tumor acting as antigen sources for T-cell expansion and systemic surveillance targeting micro-metastases.³² The residual surgical specimen can provide information on molecular and surrogate biomarkers markers that may influence the choice of subsequent personalized therapies or surveillance strategies. One such surrogate is the use of major pathological response (MPR), defined as presence of 10% or less of viable tumor in the surgical specimen after neoadjuvant chemotherapy. Using a standardized method for evaluation of tumor regression, one study found that each additional percentage point of viable tumor was significantly associated with a 1% increase in the risk of death (HR = 1.01, P = .005) and decrease in disease-free survival (HR = 1.01, P = .01).³³ This led to standardized definitions for MPR and complete pathological response (PR), to be used as surrogates in contemporary neoadjuvant clinical trials.³⁴ This contrasts with adjuvant therapy where a “one size fits all” approach is used after surgery. Due to this advantage, there has been a slew of neoadjuvant therapy trials evaluating the impact of newer agents, including checkpoint inhibitors, on MPR.

Neoadjuvant Chemotherapy in Early NSCLC

The benefits of neoadjuvant chemotherapy on improved medial survival (up to 23 months) were initially demonstrated by single-arm studies and then randomized phase II studies in patients with stage IIIA disease.^35-39 These trials were small with 27-60 patients, utilizing disparate chemotherapy regimens. They were also plagued by high risk of bias due to their small size, some being single center studies or having been stopped prematurely.^40-42 The last of the initial phase III trials was also the largest with 355 stages IB-IIIA NSCLC patients who were randomized to perioperative chemotherapy with 2 cycles of mitomycin, ifosfamide, and cisplatin followed by surgery and 2 more cycles of chemotherapy versus surgery alone. The median survival improved from 26 months to 37 months between the surgery alone arm and the perioperative chemotherapy arm.⁴³ The key common findings remained that neo-adjuvant platinum-based chemotherapy improved outcomes in operable NSCLC. However, in parallel with these trials other data emerged showing OS benefit from adjuvant chemotherapy in operable NSCLC. Subsequent planned neoadjuvant chemotherapy trials were terminated early due to the positive findings from the adjuvant trials. The NSCLC Meta-analysis Collaborative Group analyzed 15 randomized trials (2385 patients) and showed a 13% reduction in relative risk of death at 5 years, corresponding to an absolute increase in 5-year OS rate from 40% to 45%.⁴⁴ This treatment benefit was remarkably similar to the survival benefits noted with adjuvant cisplatin-based chemotherapy.^12,14 A further meta-analysis comparing 22 adjuvant chemotherapy trials and 10 neo-adjuvant trials demonstrated no significant difference in DFS or OS based on the timing of administration of chemotherapy.⁶ This has led to the addition of neoadjuvant chemotherapy as an acceptable treatment option in the NCCN guidelines.²¹ However, in clinical practice, adjuvant chemotherapy is more widely adopted, partly due to physician preference. This trend may change in the future given the positive results from recent trials as well as the recognized advantages of neo-adjuvant therapies listed above. We will next discuss the recent advances in neo-adjuvant therapy with targeted agents and ICIs.

Neoadjuvant Targeted Therapy in Early NSCLC

The successful ADAURA trial results prompted the evaluation of neoadjuvant osimertinib either alone or in addition to platinum-based chemotherapy in the ongoing NeoADAURA trial⁴⁵ Similarly, the GEOMETRY-N study is evaluating the efficacy of neo-adjuvant capmatinib in patients with MET exon 14 skipping mutations or high-level MET amplification.⁴⁶ The NAUTIKA-1 study is a multi-arm phase II neoadjuvant study evaluating multiple targetable biomarkers such as ALK, RET, NTRK fusions/ROS1, and BRAF mutations using alectinib, pralsetinib, entrectinib, and vemurafenib+cobimetinib respectively.⁴⁷ These neoadjuvant trials with targeted therapies with or without chemotherapy use MPR in the surgical specimen as the primary endpoint. The results from these trials are awaited and if positive, will expand the therapeutic armamentarium for surgically resectable NSCLC with actionable oncogenic drivers. However, it is unclear how residual viable tumors in the surgical specimen will inform adjuvant or maintenance strategies with this approach, further rigorous, biomarker-driven trials are necessary. Long-term data correlating MPR and OS are also lacking. Taken together, neoadjuvant targeted therapies in resectable NSCLC are only recommended within the context of a clinical trial.

Neoadjuvant Immune-Checkpoint Inhibitor in Early NSCLC

The Checkmate 159 study was the first published report to report the efficacy and safety of ICIs in early-stage NSCLC. This study used nivolumab for 2 doses prior to surgery and found an MPR of 45% with no delay in surgery with only one >grade 3 adverse event among the 22 patients.⁴⁸ Other studies have evaluated atezolizumab, sintilimab, and durvalumab in a similar setting.^49-51 The largest among these, the LCM3 multicentre phase II trial evaluated the efficacy and safety of 2 doses of neoadjuvant atezolizumab followed by surgery and adjuvant chemotherapy, reporting an MPR of 20% with a 3-year OS of 80%. This is impressive considering that almost half of the patients had stage III disease. However, the interpretation of these results should be considered in the context of the fact that 30% of the patients also received atezolizumab in the adjuvant setting.⁵² Taken together, these studies have demonstrated a MPR of 20%-40% with manageable toxicity profile and no significant detrimental effect on surgery or surgical outcomes. The tumor mutation burden (TMB) served as a good predictor of MPR (mutational burden among those with MPR versus those without, 311 ± 55 vs. 74 ± 60, P = .01) along with decreased FDG SUV uptake (>30% decrease in SUV-max) on pre-operative PET-CT. The MPR correlated with a concordance of T-cell clonality between tumor tissue and peripheral blood. Overall, while single ICI neoadjuvant therapy is associated with major MPR, OS data are not mature with these trials. As a result, this strategy has not found its way into the NCCN guidelines and is not recommended as standard practice. Details regarding the individual trials are presented in Table 2.

Table 2.

Prospective single-arm trials of neo-adjuvant immune check-point inhibitors as a single agent.

Trial name, sample size	Design, phase, disease stage	Intervention	Primary outcome	Results	Factors predictive of MPR	Comments
Checkmate 15948 N = 22	Single arm, phase II, Stage I-IIIA	Nivolumab 3 mg/kg × 2	Delay in surgery Grade3/4 AEs	No delay in surgery. 1 Gr3AE, MPR-45% pCR-15%	Higher TMB Tumors without PDL1 expression with infiltration of PDL1 positive ICs Concordance in T-cell clones in peripheral blood and tumor tissue had better MPR and outcomes.	The first published report on efficacy and safety of ICI in early lung cancer. Radiological response did not always correlate with pathological response PR 10% 2/21 SD 86% 18/21 PR 5% 1/21
LCM352 N = 181	Single arm, phase II, Stage I-IIIA	Atezolizumab 1200 mg q21 days × 2	Proportion with MPR	MPR-20.8% pCR 7%	MPR rate in patients with TMB <10-13% (8 of 60), 10-15—10% (1 of 10), ≥16 mutations per Mb--33% (5 of 15). Higher prevalence of non-NK cell/non-T-cell expressing immunoregulatory receptors.	Those with clinical benefits received 12 months of Atezolizumab (adjuvant). 3 year. OS 80% EGFR/ALK mutated cancers were included but analyzed separately.
Gao et al50 N = 40	Single arm, phase Ib, Stage I-IIIA	Sintilimab 200m g q21 days × 2	Grade 3 AEs Proportion with MPR ORR	10% Gr3AE 40.5% MPR 20% ORR	>30% reduction in SUV max PDL-1 expression in the stromal cells at the primary site at baseline	Decrease in SUV on post-treatment PET-CT correlated with pathological response
IONESCO51 N = 50	Single arm, phase II, Stage I-IIIA	Durvalumab 750 mg q14 days × 3	Proportion of complete (R0) resection	90% R0 resection MPR 18.6% PCR 3%		Study stopped early due to an excess of 90-day post-operative mortality. No direct durvalumab IRAE noted. *Only conference abstract available

Neo-Adjuvant Chemo-Immunotherapy

Based on the encouraging findings from single-agent ICI trials as well as the synergistic effects of chemotherapy with immunotherapy noted in the advanced NSCLC setting^7,53 the combination of platinum-based chemotherapy with ICIs as peri-operative therapy was proposed. A study by Chu et al utilizing chemotherapy and atezolizumab for 4 cycles prior to surgery, demonstrated high MPR and R0 resection rates of 57%, and 87%, respectively, with no new safety signals or increased surgical morbidity compared to historical controls.⁵⁴ The NADIM trial evaluated 2-year PFS after 3 cycles of neoadjuvant chemotherapy with nivolumab followed by adjuvant nivolumab for 1 year. The 77.1% 2-year PFS obtained with this study is impressive when we consider that they included only patients with stage IIIA disease.⁵⁵ The SAKK 14/16 trial used durvalumab with cisplatin-based chemotherapy for 2 cycles followed by adjuvant durvalumab for 1 year and demonstrated a 1-year EFS of 73% and MPR of 62%.⁵⁶ The single-arm combinations of chemotherapy and ICI confirmed that both approaches together resulted in greater response rates and PFS than seen with either modality alone. This prompted the next set of randomized trials for head-to-head comparison. The individual trial details of the above single-arm trials are presented in Table 3.

Table 3.

Prospective single-arm trials of neo-adjuvant chemotherapy + checkpoint inhibitors.

Trial name, sample size	Design, phase, disease stage	Intervention	Primary outcome	Results	Survival	Comments
Chu et al54 N = 30	Single arm, phase II, Stage Ib-IIIA	Carboplatin AUC5 D1 + nab-paclitaxel 100 mg/m2 D1,8,15 + atezolizumab 1200 mg D1 q21 days, up to 4 cycles	Proportion of MPR	57% MPR	At 12 month follow up: 19 (63%) disease-free mDFS 17·9 months (95% CI, 14·3 to not reached)	High MPR with manageable toxicity profile. 97% of patients were operated and 87% had a R0 resection. Does not impact operability PDL1 expression did not correlate with MPR or ORR.
NADIM trial55 N = 46	Single arm, Phase II, Stage IIIA	Carboplatin AUC 6 + paclitaxel 200 mg/m2 + nivolumab 360 mg q 21 days × 3 After surgery- adjuvant nivolumab × 12 months	PFS at 24 months	Gr3 AE- 30% MPR-34 (73.9%)	Two-year PFS-77.1% (95% CI, 59.9-87.7). Among those with MPR-PFS at 2 years 88.4% (67.1-96.1)	EGFR mutated and ALK translocated patients were excluded.
SAKK 16/1456 N = 68	Single arm, Phase II, Stage IIIA (N2)	Cisplatin + docetaxel q21 days × 3 followed by durvalumab 750 mg q14 days × 2. Adjuvant durvalumab × 12 months	1-year EFS	MPR 62% Gr 3 AE- 88%	73% 1-year EFS mEFS and OS not reached after 28.6 months follow-up	No significant effect of pre-treatment PDL1 expression on MPR or nodal downstaging

Neoadjuvant Chemo-Immunotherapy Versus Chemotherapy in Early NSCLC

Checkmate 816 was the first RCT of neo-adjuvant platinum-based chemotherapy with ICI (nivolumab) versus chemotherapy only. Of the 358 patients, most (60%) had stage IIIA disease. The study met its primary endpoint of improvement in event-free survival (EFS), which was defined as the time from randomization to any progression of disease precluding surgery, progression, or recurrence of disease after surgery, progression of disease in the absence of surgery, or death from any cause. The median EFS was 31.6 and 20.8 months between the experimental and control arms, respectively. This translated to a 37% improvement in EFS with the addition of 3 doses of Nivolumab to neo-adjuvant chemotherapy. This improvement in EFS was seen despite the use of adjuvant chemotherapy in 22% of the patients in the control arm, highlighting the importance of early immune priming of the tumor. A pre-planned subgroup analysis demonstrated increasing response rates along with increasing PDL-1 expression. Additionally, this study demonstrated a pathological complete response rate of 24%, which is similar to what was seen in the single-arm studies with no detrimental effects on operability.⁵⁷

The NADIM II is a randomized phase II multicentre study evaluating the efficacy of 3 cycles of carboplatin + paclitaxel and nivolumab compared to chemotherapy only for a primary endpoint of pCR. Although the final results are awaited, primary endpoint results were presented at ASCO 2022, showing a statistically significant difference in pCR at 36.2% versus 6.8% between the experimental and control arms respectively.⁵⁸ Additionally, the preliminary PFS (PFS) data were presented at the IASLC 2022 demonstrating mPFS at 24 months of 67.3% (95% CI, 55.5-81.6) for patients treated with nivolumab plus chemotherapy versus 52.6% (95% CI, 36.8- 75.2) for patients treated with chemotherapy (hazard ratio: 0.56; 95% CI, 0.28-1.15; P = .117). PDL1 expression (≥1%) identified patients who had a significant improvement in PFS (HR: 0.26; 95% CI, 0.08-0.77; P = .015) The OS at 24 months was 85.3% (95% CI, 75.7-96.1) with nivolumab plus chemotherapy versus 64.8% (95% CI, 47.4-86.4) with chemotherapy alone(hazard ratio, 0.37; 95% CI, 0.14-0.93; P = .003).⁵⁹

A meta-analysis of ICI and chemo-ICI studies have been recently published; the pooled MPR rate and pCR rate of neoadjuvant immunotherapy and chemoimmunotherapy were 43.5% and 21.9%, respectively, remarkably higher if using previously published data on neoadjuvant chemotherapy as a historical control which showed a pooled pCR rate of 4%.⁶⁰ However, the pooled MPR rates were lower than that reported with single-agent ICI studies, and pooled pCR rates were somewhat lower than those reported in the chemo-ICI studies.

Taken together, the combination of platinum-based chemotherapy and ICI could be considered as an option in operable NSCLC and is included as a recommendation in the most recent NCCN guidelines.²¹ The individual trial details of the randomized trials of neo-adjuvant chemo-immunotherapy or dual checkpoint blockade trials are presented in Table 4. The individual trial details for ongoing trials of neo-adjuvant immunotherapy and/or chemo-immunotherapy in early NSCLC are presented in Table 5.

Table 4.

Prospective randomized controlled trials of neo-adjuvant chemotherapy + immunotherapy OR dual-check-point blockade.

Trial name, sample size	Design, phase, disease stage	Intervention	Primary outcome	Results	Survival	Comments
NEOSTAR61 N = 44	Randomized controlled trial, Phase II, Stage Ib-IIIA	Nivolumab 3 mg/kg (N) Q14 days × 3 vs. Nivolumab 3 mg/kg Q14 days x 3+ opilimumab (NI) 1 mg/kg × 1	Proportion of MPR	N = 22% MPR NI = 38% MPR	Median RFS and OS not reached. Numerically fewer events among those with MPR (1/13, 8%) versus those without MPR (3/24, 13%).	89% resection rate overall. N + I had less viable tumor (median 50% vs. 9%) and greater tissue resident memory and effector memory T cells.
Checkmate 81657 N = 773	Randomized controlled trial, Phase III, Stage Ib-IIIA	Nivolumab + platinum-based chemotherapy × 3 versus Platinum-based chemotherapy	EFS PCR	MPR-36.9% versus 8.9% pCR = 24.0% (95% CI, 18.0-31.0) and 2.2% (95% CI, 0.6-5.6)	mEFS 31.6 months versus 20.8 months HR for EFS-0.63; 97.38% CI, 0.43-0.91; P = .005	Trend toward longer EFS among those with pCR vs. those with residual tumor 26.6 vs. 18.4 months HR 0.84; 95% CI, 0.61-1.17 Nivo + chemo was beneficial irrespective of PDL1 status-higher PDL1 benefitted more. Higher ctDNA clearance was noted among those who achieved PCR.
NADIM II58,59	Phase II	Nivolumab 360 mg + carboplatin AUC5 + paclitaxel q21 days × 3 followed by adjuvant nivolumab × 6 months versus carboplatin AUC5 + paclitaxel q21 days × 3	pCR	36.2% vs. 6.8% (RR = 5.25 [99% CI, 1.32-20.87]; P = .0071)	Not reported	Grades 3-4-related AEs were reported in 24 % of Nivo + chemo versus 10% of chemo alone. *Only preliminary results are available as abstracts.

Table 5.

Ongoing trials of neo-adjuvant immunotherapy and/or chemo-immunotherapy in early lung cancer.

Trial name, sample size	Design, phase, disease stage	Intervention	Primary outcome	Results	Survival	Comments
PRINCEPS62	Single arm, phase II	Atezolizumab 1200 mg × 1	Response by RECIST 1.1 Proportion of major pathological response	No RECIST response No MPR		Only single dose of ICI was given. 15 (50%) were stage I. More time and/or more doses of IO are needed for MPR
TOP 1501 NCT0281892063	Single arm, phase II	Pembrolizumab 200 mg q21 days × 2 Standard adjuvant chemotherapy after surgery	Surgical feasibility rate	Awaited	Awaited
Keynote 671 NCT0342564364	Randomized controlled trial, phase III	Cisplatin-based chemotherapy + Pembrolizumab 200 mg q21 days × 4 followed by surgery and adjuvant Pembrolizumab × 13 versus Cisplatin-based chemotherapy × 4 followed by surgery	EFS	Awaited	Awaited
AEGEAN NCT0380013465	RCT, Phase III	Cisplatin-based chemotherapy + durvalumab q21 days × 4 followed by surgery and adjuvant durvalumab × 13 versus cisplatin-based chemotherapy × 4 followed by surgery	pCR EFS	Awaited	Awaited
Impower 030 NCT0345606366	RCT, phase III	Atezolizumab + platinum-based neoadjuvant chemotherapy followed by surgery followed by adjuvant atezolizumab 16 cycles versus Placebo + platinum-based chemotherapy followed by surgery followed by best supportive care	EFS	Awaited	Awaited

Neoadjuvant Dual Immune Checkpoint Blocking Therapy in ENSCLC

ICIs targeting programmed death-1 (PD-1) receptor and it is ligand (PDL-1) work through distinct, non-redundant pathways relative to ICIs targeting cytotoxic T-lymphocyte antigen-4 (CTLA-4), prompting efforts to target both immune check-points concurrently.⁶⁷ The NEOSTAR trial randomized 44 patients with stage I-IIIA to neo-adjuvant nivolumab and ipilimumab (NI) versus nivolumab (N) only. Participants in the experimental arm received 3 cycles of nivolumab with one dose of ipilimumab followed by surgery. The study achieved its primary endpoint, showing a significant difference in MPR (38% vs. 22%) between the NI and N arms, respectively⁶¹ The pitfalls of cross-trial comparisons notwithstanding, it is important to note that the MPR rates from this trial is numerically similar to that of the Checkmate 816 trial. However, there were 9 serious adverse events (5 treatment-related) in the Neostar study of combined ICIs; the checkmate 816 had a third arm with Nivolumab and Ipilimumab which was stopped early, due to toxicities; a smaller phase II study of the combination of nivolumab and ipilimumab was discontinued early due to 67% grade 3-5 TRAEs.⁶⁸ At the current time, dual ICI as neoadjuvant therapy is not recommended; further studies of predictive biomarkers may identify who might benefit most from this approach.

Comparison of Approaches—Can We Choose?

Currently, in clinical practice, most practicing oncologists prescribe adjuvant chemotherapy for patients with stage II and III NSCLC (and some stage IB NSCLCs associated with poor prognostic factors); this adjuvant chemotherapy consists of a platinum-based doublet for 4-6 cycles. Given the ADAURA data, many practicing oncologists prescribe 2 years of Osimertinib in mEGFR lung adenocarcinomas; others await OS data. Notably, the ADAURA trial did not mandate the use of adjuvant chemotherapy prior to osimertanib. Given the IMPOWER010 data, there are practioners who prescribe atezolizumab for 1 year in patients whose lung cancers express PD-L1 > 1%, while others wait for OS data. The preferential practice of adjuvant chemotherapy/targeted or immunotherapy may be attributed to the fact that the timing does not affect “curative” surgical resection.

The shift in clinical practice to a neoadjuvant approach requires a discussion of key questions:

• (i) Is there a sound biological rationale for ICI-based neoadjuvant approaches? The theoretical superiority of neoadjuvant ICI in cancers was initially demonstrated in pre-clinical mouse models of metastatic breast cancer.⁶⁹ Recently, this theory found further support from randomized trials comparing neoadjuvant and adjuvant pembrolizumab in resected melanoma.^70,71 The checkmate 159 trial also demonstrated that concordance of T-cell clones between the peripheral blood and the primary was associated with greater tumor regression and MPR after nivolumab.⁴⁸ Biologically, an intact tumor and lymphatics may potentially afford a greater benefit from chemo-ICI-based approaches in NSCLC as cytotoxic chemotherapy can result in tumor cell kill which releases more neoantigens that result in greater priming of tumor-specific T cells (Fig. 1); these expanded T-cell clones may facilitate both improved PFS and OS by targeting immune evasion in the short- and longer-term (immunological memory).

• (ii) Can PCR or MPR be a useful surrogate of PFS and OS? Checkmate 816 has shown that an improved PCR was associated with an improvement in EFS (HR 0.84 [0.61-1.17]).⁵⁷ The NEOSTAR trial demonstrated a numerically lower EFS rate among those achieving MPR (1/13, 8%) versus those without MPR (3/24, 13%).⁶¹ Taken together, MPR and pCR appear to be good surrogates for EFS, but whether they represent good surrogates of OS remains to be validated. Proponents of OS as the only valid endpoint for neoadjuvant studies consider the potential toxicities that patients with earlier stage NSCLCs (particularly stages I and II NSCLC) may be needlessly exposed to. Opponents cite the argument that availability of effective therapies in the advanced disease setting makes demonstrating an OS benefit more difficult (from neoadjuvant approaches) due to a dilution of the benefits from subsequent interventions.

• (iii) Can neoadjuvant therapy be delivered without compromising time to surgical resection? During the earlier neo-adjuvant chemotherapy trials approximately 20% of patients did not undergo surgery⁵ thus contributing to the poor uptake of this option in clinical practice. The surgical outcomes from Checkmate 816, report that 7.8% (14/179) of those in the nivolumab + chemotherapy had canceled surgery due to disease progression or adverse events as compared to 10.1% (18/179) in the control arm. Furthermore, 6 patients (4%) and 9 patients (6.7%) had a delay of >6 weeks from the last dose of neoadjuvant therapy to surgery due to an adverse event in the experimental and control arms, respectively. The NADIM II trial presentation at ASCO 2022 reported a higher proportion of patients who underwent surgery in the experimental arm as compared to the control arm.^57,58 Based on the available data, the addition of ICI to neoadjuvant chemotherapy does not adversely impact operability. Furthermore, there were no significant differences in progressive cancer between the interventional and control arms of these studies.

• (Iv) Will a neoadjuvant approach result in more surgical morbidity? The surgical outcomes from the Checkmate 816 and NADIM II demonstrate low attrition rates with no difference in operative approach, blood loss, or post operative complications with an overall trend favoring the Nivolumab arm.^57,72 This is in keeping with the findings from other phases II studies of ICI therapy in operable NSCLC that do not show a detrimental effect on operability or surgical outcomes.^48,54,61

• (v) What effect would demonstration of viable residual tumor following neoadjuvant chemo-ICI or targeted therapies have on the use of adjuvant ICI in patients with PD-L1 > 1%? There is currently no data to guide practice in this scenario. Rigorous trials that randomize patients to further ICI or placebo may provide definitive answers. Specifically, blood-based biomarkers, such as circulating tumor DNA (ctDNA) may inform adjuvant strategies. This was already explored in Checkmate 816.⁵⁷ EFS was longer in patients with ctDNA clearance than in those without in both the nivolumab-plus-chemotherapy group and the chemotherapy-alone group The percentage of patients with a pCR was higher among those with compared to no ctDNA clearance. Furthermore, the investigators in the NADIMII study showed that undetectable ctDNA levels after neoadjuvant treatment were significantly associated with improved PFS and OS.⁷³ Taken together, ctDNA following neoadjuvant therapy and surgery may inform adjuvant approaches, including surveillance.

• (vi) Could an approach using concurrent chemo-XRT prior to surgical resection (especially in resectable stage IIIA) followed by adjuvant ICI have the same outcome as the intervention arm described in the checkmate 816 trial? Current clinical practice and NCCN guidelines recommend consideration of chemotherapy or CRT in resectable stage IIIA, including Pancoast tumors prior to curative surgery.²¹ Furthermore, such an approach could incorporate ICI in the adjuvant setting, in the event of residual viable cancer in the surgical specimen. It is unknown whether such an approach would have closed the OS gap demonstrated in the Checkmate 816 study.

• (vii) Finally, would a neoadjuvant chemo-ICI approach have the same OS outcomes as adjuvant chemo-ICI in resectable NSCLC? Prior studies of neoadjuvant versus adjuvant chemotherapy approaches in stage IIIA did not show a statistically significant difference in OS.⁶ It may be the case that in resectable stage IIIA disease, a similar, equivalent outcome for OS may be seen whether chemo-ICI is used as neoadjuvant versus adjuvant treatment, particularly in patients whose tumors express PD-L1 50%. Such a study, while desirable to definitively answer the question may not be feasible given the more pressing questions in the field that will compete for resources and patients.

Figure 1.

Two potential mechanisms for the enhancement of systemic antitumor T-cell immunity after neoadjuvant PD-(L)1 blockade. PD-(L)1 blockade could result in the “in situ” expansion of tumor-specific T-cell clones already within the tumor microenvironment. This expansion and activation are largely driven by PD-L1− and PD-L2− expressing dendritic cells in the tumor. Tumor-specific tumor-infiltrating lymphocytes may represent naïve T cells or T cells that have already been “primed” to tumor antigen before PD-1 pathway blockade. In addition, tumor antigen-containing dendritic cells that originate in the tumor pick up tumor antigens and traffic to the tumor-draining lymph nodes, where they present antigens either ineffectively or in a tolerogenic fashion to tumor-specific T cells. PD-(L)1 blockade could act at this point, enhancing productive stimulation of tumor-specific T cells or partially reversing tolerance induction. Activated T cells enter the circulation by way of efferent lymphatics and then egress into tissues. Topalian et al: Neoadjuvant checkpoint blockade for cancer immunotherapy, Volume: 367, Issue: 6477, DOI: (10.1126/science.aax0182) Permission License # 5486050581126.

So, how do we choose? Neoadjuvant ICI and chemo-immunotherapy result in high MPR and pCR rates with tolerable toxicity profiles across all histologies of NSCLC, with the most benefit in PD-L1 expressing NSCLC. However, the effects on long-term survival and recurrence are yet to be determined. In terms of which stages benefit, it may be the same stages that benefit from adjuvant chemotherapy. Interestingly, the same group of patients are now eligible for adjuvant ICI with atezolizumab for 1-year. At the current time, neoadjuvant chemo-ICI is another arrow in the quiver of effective therapies for resectable NSCLC, with the most clinical rationale and clinician buy-in for resectable stage IIIA NSCLCs with PD-L1 expression of >1%. As OS data mature, there may be more adoption of this approach. Window of opportunity studies incorporating novel ICI combinations or targeted therapy-ICI combinations should be conducted to rigorously evaluate the impact of newer biologics or combinations on residual tumors.

Future Directions

Systemic therapy, either neoadjuvant or adjuvant is an essential part of therapy for larger and lymph node positive NSCLC, much like what is recommended for breast cancer. Identifying patients who are most likely to benefit from neoadjuvant approaches will require a combinatorial contemporary approach such as validated machine learning models that takes into consideration myriad data points that predict who might recur after curative surgery. On the other hand, predictive markers might also identify early NSCLCs that might be cured with systemic therapy alone using ICI-based approaches in select patients, foregoing surgery altogether or reserving this as a salvage option. The surgical specimen and post-operative blood-based biomarkers, such as ctDNA may inform who might require adjuvant therapy versus surveillance, which may also be used to escalate therapy, during follow-up.

Supplementary Material

Supplementary material is available at The Oncologist online.

Funding

Funded by Veterans Affairs Ann Arbor Lung Precision Oncology Program (VA I50CU000182).

Conflict of Interest

The authors indicated no conflict of interest.

Author Contributions

Both authors contributed to the conception/design, manuscript writing, and final approval of manuscript.

Data Availability

No new data were generated or analyzed in support of this research.