How Do Immune Checkpoint Inhibitors Work?

ICIs allow for T cells to mount an antitumor response by overcoming the normal regulatory mechanisms of T cell activation. Tumor cells express tumor-specific antigens in the context of the major histocompatibility complex, which are presented on antigen-presenting cells (APCs), allowing the T cell to recognize the tumor. A second signal is needed to activate the T cell, which involves the CD28 receptor on T cells engaging with the B7 receptor (CD80/86) on the APC. T cell activation is actually more complex, however, as it also induces an inhibitory pathway that provides self-regulation by either attenuating or abrogating the T cell response. T cells express immune checkpoint molecules, such as cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and programmed cell death protein 1 (PD-1), which are upregulated on T cell activation. CTLA-4 downregulates T cell activation by outcompeting CD28 for B7 (CD80/86) and inducing T cell cycle arrest, thus decreasing T cell activation. ^, The hypothesis that an antibody against CTLA-4 would block its interaction with B7 and consequently allow sufficient T cell responses to increase antitumor activity was demonstrated in mice and led to the development of ipilimumab, which ultimately because the first FDA-approved ICI granted for the treatment of metastatic melanoma.

After the clinical success of CTLA-4, other immune checkpoint molecules were targeted, including PD-1. PD-1 is expressed on several immunologic cells, including B cells, natural killer cells, and monocytes, and, unlike CTLA-4, directly regulates T cell activation. Activation of T and B lymphocytes leads to expression of PD-1, which binds to its ligands PD-L1 and PD-L2. PD-L1 and PD-L2 are expressed in response to inflammatory cytokines, and when PD-1 binds to them it inhibits the kinase signaling pathway that ordinarily activates T cells. This helps protect against autoimmunity in the normal immune milieu, but also allows for tumors to evade the immune system, as PD-L1 and PD-L2 can both be expressed on tumor cells. Blockade of PD-1 or its ligand PD-L1 removes this negative signal and, as the ligands are widely expressed in nonlymphoid tissues, restores the antitumor T cell response in the periphery ( Fig. 1 ).

Mechanism of action of immune checkpoint inhibitors.

Nivolumab

Nivolumab, a fully human immunoglobulin G4 (IgG4) monoclonal antibody against PD-1, was the first ICI granted FDA approval for chemotherapy-treated NSCLC. It was approved in 2015 based on a randomized, open-label, international, phase III study (CheckMate-017) comparing nivolumab versus docetaxel in previously treated patients with advanced squamous NSCLC (n = 272). At the pre-planned interim analysis, overall survival (OS), 1-year OS, and objective response rate (ORR) all statistically significantly favored nivolumab, leading to early closure of the trial. Median OS was 9.2 months in the nivolumab group (95% CI: 7.3–13.3) versus 6.0 months in the docetaxel group (95% CI: 5.1–7.3); hazard ratio (HR) was 0.59 (95% CI: 0.44–0.79; P <.001); and the 1-year OS with nivolumab of 42% was significantly higher than with docetaxel of 24%. In addition, progression-free survival (PFS) at 1 year was 21% (95% CI: 14–28) in nivolumab and 6% (95% CI: 3–12) in docetaxel. Treatment-related adverse events (AEs) were less frequent in the nivolumab group (58% of any grade) than the docetaxel group (86% of any grade), and led to treatment discontinuation less frequently in nivolumab (3% of patients) than in docetaxel (10% of patients), as per Table 1 .

Nivolumab was concurrently evaluated in patients with previously treated, advanced, non-squamous NSCLC (n = 582) in a randomized, open-label, international, phase III study (CheckMate-057), again compared with standard second-line docetaxel. The pre-planned interim analysis revealed statistical significance in OS, 1-year OS, and ORR in favor of nivolumab, which led to early closure of the trial. Median OS was 13.2 months in the nivolumab group (95% CI: 9.7–10.5) and 9.4 months in the docetaxel group (95% CI: 8.1–10.7), with HR of 0.73 (95% CI: 0.59–0.89, P = .002). The 1-year OS for nivolumab was 51% compared with 39% for docetaxel. ORR for nivolumab of 19% was significantly higher than for docetaxel of 12%. Although median PFS for nivolumab was 2.3 months versus 4.2 months for docetaxel, the 1-year PFS for nivolumab was 19% versus 8% for docetaxel. There were fewer grade 3+ AEs in the nivolumab group (10%) than in the docetaxel group (54%). Discontinuation of treatment was less frequent with nivolumab (5%) than with docetaxel (15%). These results led to expansion of FDA approval for second-line monotherapy with nivolumab for advanced non-squamous NSCLC (see Table 1 ).

Pembrolizumab

Pembrolizumab is a humanized IgG4 monoclonal antibody against PD-1, which was first evaluated in a phase 1 trial in advanced melanoma. In this study, pembrolizumab monotherapy demonstrated a 38% ORR and overall median PFS that was not greater than 7 months. In NSCLC a subsequent international phase 1 trial (Keynote-001) evaluated pembrolizumab in both previously treated and previously untreated patients (n = 495) at 3 different dose levels: 2 mg/kg every 3 weeks, 10 mg/kg every 3 weeks, or 10 mg/kg every 2 weeks. ORR was similar across dose, schedule, and tumor histology analyses, with an ORR of 19.4% (95% CI: 16.0–23.2). This included an 18.0% response rate in previously treated patients and 24.8% response rate in previously untreated patients. Median duration of response (DOR) was 12.5 months for all patients, with 84.4% of patients who had a response without disease progression. Median OS was 12.0 months for all patients. The trial also evaluated PD-L1 immunohistochemical (IHC) expression as a potential predictive biomarker of response, and in patients with PD-L1 expression of at least 50%, ORR was 45.2% (95% CI: 33.5–57.3). As a result, not only did pembrolizumab receive FDA approval as second-line monotherapy treatment of PD-L1-positive advanced NSCLC, but the companion diagnostic PD-L1 IHC 22C3 pharmDx test was also approved.

The success of Keynote-001 led to a subsequent phase II/III trial (Keynote-010), which evaluated pembrolizumab in patients with previously treated PD-L1-positive (tumor proportion score greater than 1%) NSCLC (n = 1034) at 2 doses, 2 or 10 mg/kg, versus docetaxel. The primary endpoints of OS and PFS were evaluated, with a concurrent goal to validate the 22C3 biomarker with a PD-L1 cutoff score of ≥50%. In the total population, median OS was significantly improved for both doses of pembrolizumab at 10.4 months in the 2-mg/kg group (95% CI: 9.4–11.9), 12.7 months in the 10 mg/kg group (95% CI: 10.0–17.3), and 8.5 months versus docetaxel. In the patients with ≥50% PD-L1 expression, PFS was significantly longer with HR of 0.59 ( P = .0001) for pembrolizumab 2 mg/kg versus docetaxel, and also HR of 0.59 ( P <.0001) for pembrolizumab 10 mg/kg versus docetaxel. More impressively, median OS for the group of patients with greater than 50% PD-L1 expression was 14.9 months in the pembrolizumab 2 mg/kg group (95% CI: 10.4–not reached), 17.3 months in the pembrolizumab 10 mg/kg group (95% CI: 11.8–not reached), and 8.2 months in the docetaxel group (see Table 1 ). Grade 3+ treatment-related AEs were highest in the docetaxel group at 35% compared with 13% in the pembrolizumab 2 mg/kg group and 16% in the pembrolizumab 10 mg/kg group.

These data also supported the investigation of pembrolizumab monotherapy in a first-line phase III trial of patients with metastatic NSCLC (n = 305) with PD-L1 expression ≥50% (Keynote-024) comparing pembrolizumab (200 mg fixed dose) to platinum chemotherapy of the investigator’s choice. In this practice-changing study, the median PFS was 10.3 months (95% CI: 6.7–not reached) in the pembrolizumab group, and 6.0 months in the chemotherapy group, with HR of 0.50 (95% CI: 0.37–0.68, P <.001). OS was significantly prolonged with an HR of 0.60 (95% CI: 0.41–0.89, P = .005), and estimated OS at 6 months 80.2% with pembrolizumab versus 72.4% with chemotherapy. This led to early closure of the trial to allow patients in the chemotherapy arm to receive pembrolizumab, as well as FDA approval for pembrolizumab as first-line monotherapy in patients with PD-L1 expression ≥50%. Most recently, the FDA broadened approval for pembrolizumab in the first-line setting to include patients with stage III NSCLC and also lowered the PD-L1 cutoff to ≥1% based on results from Keynote-042. In this larger phase III trial comparing pembrolizumab (200 mg) versus platinum-based chemotherapy in patients with previously untreated advanced NSCLC (n = 1274) at various PD-L1 expression levels, median OS was significantly improved in all of the pembrolizumab groups (see Table 1 ). These results confirmed efficacy of pembrolizumab monotherapy in patients with PD-L1 ≥50% and also demonstrated benefit in patients with the lower cutoff of PD-L1 ≥1%.

Atezolizumab

Atezolizumab is a humanized IgG1 monoclonal antibody against PD-L1 that was FDA approved for the treatment of patients with metastatic NSCLC in the second-line setting, based on positive results in large phase I, phase II (POPLAR), and phase III (OAK) trials. The OAK trial was a large, randomized, international trial comparing atezolizumab (1200 mg) with docetaxel in patients with advanced NSCLC (n-1225), stratified by PD-L1 expression: PD-L1 ≥1%, PD-L1 ≥5%, and PD-L1 ≥50%. Median OS in the intention to treat population was statistically significantly greater in the atezolizumab group at 13.8 months (95% CI: 11.8–15.7) compared with 9.6 months in the docetaxel group (95% CI: 8.6–11.2). OS was improved with atezolizumab regardless of level of PD-L1 expression, with mOS in all PD-L1-positive patients treated with atezolizumab of 15.7 months (95% CI: 12.6–18.0) versus 10.3 months with docetaxel (95% CI: 8.8–12.0). PFS was similar for patients treated with atezolizumab (2.8 months) versus docetaxel (4.0 months), but median DOR was longer with atezolizumab (16.3 months) compared with docetaxel (6.2 months). There were more treatment-related AEs leading to discontinuation in the docetaxel group (19%) than in the atezolizumab group (8%).

Durvalumab

Durvalumab is a high-affinity, human IgG1 monoclonal antibody against PD-L1, which was first FDA approved for the treatment of previously treated locally advanced or metastatic urothelial carcinoma based on the interim analysis of a phase I/II trial demonstrating efficacy in the PD-L1-positive subgroup. Following encouraging results in a phase I/II dose-escalation and expansion study, durvalumab was evaluated in a randomized, international phase III trial (PACIFIC) as maintenance therapy after chemoradiotherapy for stage III, unresectable NSCLC. Patients received either durvalumab 10 mg/kg as consolidation therapy for up to 12 months versus placebo (n = 709). The planned interim analysis met its primary endpoint of PFS: the median PFS from randomization was 16.8 months with durvalumab (95% CI: 13.0–18.1) versus 5.6 months with placebo. The 18-month PFS rate was 44.2% for durvalumab (95% CI: 37.7–50.5) compared with 27.0% with placebo, and this PFS benefit was irrespective of PD-L1 expression. AEs were similar in both groups, and grade 3 or 4 AEs were 29.9% with durvalumab compared with 26.1% with placebo. These results led to FDA approval of durvalumab as consolidation therapy, before reporting of the second primary endpoint of OS, which was impressive as median OS was not reached in the durvalumab group with 33.3 months median follow-up, compared with mOS 29.1 months in the placebo group. Updated PFS was also reported, with median PFS with durvalumab 17.2 months (95% CI: 13.1–23.9) versus 5.6 months (95% CI: 4.6–7.7) with placebo. The safety profile was similar, and again noted was the higher rate of pneumonitis with durvalumab compared with placebo. This was further explored, and both Asian patients (47.9% vs 17.6%) and patients with EGFR mutations (11.0% vs 3.8%) were more likely to have AE pneumonitis.

Ipilimumab Plus Nivolumab

Ipilimumab is a fully human anti-CTLA-4 that was first approved by the FDA as monotherapy for previously treated metastatic melanoma based on phase III data. The combination of ipilimumab and nivolumab was then evaluated in untreated melanoma in a phase I trial evaluating both sequential and concurrent treatment. Positive results in the ensuing phase II trial comparing the combination of ipilimumab plus nivolumab versus ipilimumab monotherapy led to FDA approval of the combination for the treatment of unresectable or metastatic melanoma. Given the evidence in favor of combination immunotherapy for melanoma, the combination of ipilimumab plus nivolumab was evaluated in NSCLC in a phase 1 multicohort study (CheckMate-012). Patients were randomized to 3 different dosing/scheduling regimens: nivolumab 1 mg/kg every 2 weeks plus ipilimumab 1 mg/kg every 6 weeks, nivolumab 3 mg/kg every 2 weeks plus ipilimumab 1 mg/kg every 12 weeks, and nivolumab 3 mg/g every 2 weeks plus ipilimumab 1 mg/kg every 6 weeks, with results reported for the nivolumab 3 mg/kg doses. Safety endpoints were met with treatment-related serious AEs in 32% of the patients in the ipilimumab every 12 weeks group, and 28% of the patients in the ipilimumab every 6 weeks group. Confirmed objective responses were seen in 47% of patients treated with the ipilimumab every 12 weeks, and in 38% of the patients treated with ipilimumab every 6 weeks. This was further improved in patients with PD-L1 expression ≥1% to 57% with objective responses in both cohorts. Median PFS in the PD-L1-positive patients treated with ipilimumab every 12 weeks was 8.1 months (95% CI: 5.6–not reached), and 10.6 months (95% CI: 3.5–not reached) in those treated with ipilimumab every 6 weeks.

With these promising results, a large, randomized phase III trial (CheckMate-227) was done to further evaluate the clinical benefit of the combination of nivolumab plus ipilimumab in patients with advanced, previously untreated NSCLC (n = 1739). This 2-part trial randomized patients with PD-L1 ≥1% in part 1a to receive nivolumab plus ipilimumab, nivolumab alone, or platinum-doublet chemotherapy; patients with PD-L1 ≤1% in part 1b received nivolumab plus ipilimumab, nivolumab plus platinum-doublet chemotherapy, or platinum-doublet chemotherapy. The most notable results were in the comparison of patients treated with nivolumab plus ipilimumab compared with chemotherapy. In patients with PD-L1 ≥1%, median OS was 17.1 months (95% CI: 15.0–20.1) with nivolumab plus ipilimumab, compared with 14.9 months with chemotherapy alone ( P =.007), and the HR for death was 0.79 (97.72% CI: 0.65–0.96). More significantly, the median DOR was 23.2 months in the combination immunotherapy group versus 6.2 months in the chemotherapy group, and patients with an ongoing response at 2 years was 49.5% in the combination group versus 11.0% in the chemotherapy group. Median OS was also significantly improved in patients with PD-L1 ≤1% who received combination therapy (17.2 months, 95% CI: 12.8–22.0) compared with chemotherapy (12.2 months), with HR for death 0.62 (95% CI: 0.48–0.78). Benefit was also seen in PFS, ORR, and DOR regardless of PD-L1 expression. Although combination immunotherapy is not yet approved for the treatment of NSCLC, it is likely that it will soon be another FDA-approved indication, potentially regardless of PD-L1 expression.

An earlier phase II trial, CheckMate-568, evaluated the combination of nivolumab plus ipilimumab based on tumor mutational burden (TMB), the number of mutations in cancer cells. As it demonstrated improved response and PFS in patients with TMB ≥ 10 mutations/megabase (mut/Mb), PFS in patients with TMB ≥ 10 mut/Mb regardless of PD-L1 expression was also separately analyzed as a coprimary endpoint in CheckMate-227. Results were promising, with median PFS in the high TMB group 7.2 months with nivolumab plus ipilimumab (95% CI: 5.5–13.2) versus 5.5 months with chemotherapy, and response rate of 45.3% with nivolumab plus ipilimumab compared with 26.9% with chemotherapy. However, the OS was similar regardless of TMB and when combining PD-L1 expression and TMB. There remains no predictive biomarker of response for ICIs, which will be further discussed below (see Future Directions) .

Pembrolizumab Plus Chemotherapy

With the success of pembrolizumab as a monotherapy and more understanding of the antitumor mechanism of chemotherapy, evaluating for synergy in the combination of ICI with chemotherapy seemed like a natural next step. The addition of pembrolizumab to the platinum doublet of carboplatin and pemetrexed was first evaluated in a randomized, phase II study (Keynote-021) in patients with untreated, advanced, non-squamous NSCLC (n = 123). A significantly higher proportion of patients treated with pembrolizumab plus chemotherapy (55%, 95% CI: 42–68) had an objective response compared with the chemotherapy-only group (29%, 95% CI: 18–41). The secondary endpoint of PFS was significantly higher for patients treated with pembrolizumab plus chemotherapy (mPFS 13.0 months, 95% CI: 8.3–not reached) compared with chemotherapy alone (mPFS 8.9 months). These results led to accelerated FDA approval of the combination of pembrolizumab with pemetrexed and platinum as first-line therapy for metastatic, non-squamous NSCLC. Results from the ensuing phase III study (Keynote-189) led to regular approval the following year. Keynote-189 compared pembrolizumab plus chemotherapy or placebo plus chemotherapy in 616 patients with metastatic, untreated, non-squamous NSCLC. Both primary endpoints of OS and PFS were met, with median OS not reached in the pembrolizumab plus chemotherapy group and 11.3 months in the placebo-chemotherapy group. The HR for death was 0.49 (95% CI: 0.38–0.64). Median PFS was 8.8 months (95% CI: 7.6–9.2) in the pembrolizumab plus chemotherapy group compared with 4.9 months in the placebo-combination group. This combination regimen now constitutes a new standard of care for newly diagnosed patients with advanced non-squamous NSCLC, regardless of PD-L1 expression. The question of whether to use pembrolizumab monotherapy or the combination with chemotherapy for high PD-L1 expressors (>50%) is an open question.

Pembrolizumab in combination with chemotherapy was also evaluated in patients with untreated, metastatic, squamous NSCLC (n = 559) in the phase III Keynote-407 trial. Patients were stratified by PD-L1 expression, and the trial had 2 primary endpoints of OS and PFS, both of which were met. The median OS in the pembrolizumab plus chemotherapy group was 15.9 months (95% CI: 13.2–not reached) compared with 11.3 months in the placebo plus chemotherapy group. Benefit was seen regardless of PD-L1 expression for OS and also for PFS, with incremental improvement with increasing PD-L1 expression. The median PFS for the pembrolizumab plus chemotherapy group was 6.4 months (95% CI: 6.2–8.3) compared with 4.8 months for the placebo plus chemotherapy group. The rate of AEs was similar in both groups. The results of this trial led to FDA approval of pembrolizumab plus chemotherapy for metastatic squamous NSCLC. This is also a new recent standard of care for patients with metastatic squamous NSCLC, regardless of PD-L1 expression, with similar questions relating to choice of monotherapy versus combination in high expressors of PD-L1.

Atezolizumab Plus Chemotherapy Plus Bevacizumab

Yet another combination strategy that was recently evaluated in patients with metastatic NSCLC was the combination of atezolizumab with bevacizumab, an inhibitor of vascular endothelial growth factor was studied in the IMpower150 study. ^, IMpower150 was a large, randomized trial of patients with metastatic non-squamous NSCLC (n = 1045) comparing atezolizumab plus bevacizumab plus chemotherapy (ABCP) versus atezolizumab plus chemotherapy (ACP) versus bevacizumab plus chemotherapy (BCP). Due to improvement in OS and PFS with ABCP (mOS 19.2 months, mPFS 8.3 months) compared with BCP (mOS 14.7 months, mPFS 6.8 months) regardless of PD-L1 expression and EGFR or ALK gene alteration status, the 4-drug regimen received FDA approval. This also represents a first-line treatment option for patients with newly diagnosed stage IV NSCLC regardless of PD-L1 expression.

Biomarkers of Response

Although ICIs have transformed outcomes for patients with many other types of cancer, most NSCLC patients’ cancers will ultimately progress, and some patient’s NSCLCs will not respond to immunotherapy. This is true across tumor types, and there is currently no biomarker that has been validated across tumor types to predict response to therapy. PD-L1 expression, tested by IHC, is currently the only validated biomarker, and FDA approvals for some ICIs in several cancers, including bladder cancer, NSCLC, triple-negative breast cancer, cervical cancer, and gastric cancer are linked to PD-L1 status. There are currently 5 PD-L1 assays with FDA approval, each developed for a different ICI: PD-L1 IHC 22C3 pharmDx as a companion diagnostic for pembrolizumab, PD-L1 IHC 28 to 8 pharmDx as a complementary diagnostic for nivolumab, VENTANA PD-L1 (SP142) assay as a complementary diagnostic for atezolizumab, VENTANA PD-L1 (SP263) assay as a complementary diagnostic for durvalumab (bladder cancer only), and most recently PD-L1 IHC 73-10 assay for avelumab. The multitude of assays antibodies and different scoring systems complicates the use of PD-L1 expression in clinical practice. The Blueprint PD-L1 Assay IHC Comparison Project initially compared the first 4 assays and cutoffs, and found that the 22C3, 28-8, and SP263 assays were interchangeable, and the SP142 assay less sensitive than the others. Phase 2 of the Blueprint study also analyzed the 73-10 assay, and found it to have greater sensitivity than the other 4 assays. Phase 2B of the Blueprint study compared samples of tumor resection, small biopsy, and needle aspirate of the same tumor and found no significant difference on PD-L1 scoring among them, although scoring was not possible in 14% of the aspirates. This issue of obtaining samples for PD-L1 testing is potentially magnified in NSCLC, as cytology is a common way of diagnosing NSCLC. In addition, PD-L1 expression is heterogeneous, with low interobserver reproducibility. Although PD-L1 expression is actively used in the clinic to determine eligibility for treatment of patients with NSCLC with pembrolizumab, methods to standardize PD-L1 evaluation are ongoing.

TMB is another potential marker that has shown particular promise in NSCLC but needs to be validated prospectively. ^, Somatic mutations in the DNA of affected cells accumulate over time and ultimately cause neoplastic transformation. Some of these mutations generate neoantigens, which can be recognized by the immune system, and TMB can be used to estimate the tumor neoantigen load. Higher numbers of mutations was first found to be associated with improved response and PFS in patients with NSCLC treated with pembrolizumab retrospectively,. This led to the evaluation of TMB as a potential biomarker of response in relation to other anti-PD-1/PD-L1 ICIs. In CheckMate-026, a phase III trial of nivolumab monotherapy versus chemotherapy in patients with metastatic NSCLC with PD-L1 ≥1%, patients with high TMB (defined as ≥243 missense mutations, or the upper tertile) were found to have a higher response rate (47% vs 28%) and PFS (9.7 vs 5.8 months). Blood-based TMB (bTMB) has also been evaluated prospectively. In the B-F1RST study, a trial of atezolizumab monotherapy in patients with untreated stage III/IV NSCLC, patients with high bTMB, defined as ≥16 mut/Mb, had improved PFS and OS with atezolizumab. However, there are also severl assays for TMB, and no standardized cutoff for high TMB. In addition, the number of mutations in different tumor types is highly variable, particularly in NSCLC, as there are more mutations in the tumors of smokers than never-smokers, and not all mutations result in a neoantigen or functional T cell response. Identification of further predictive biomarkers of response to anti-PD-1/PD-L1 remains an area of active research in NSCLC and other tumor types.

The Gut Microbiome

The gut microbiome is another emerging biomarker of response that is uniquely a modifiable feature of the host, whereas TMB and PD-L1 are nonmodifiable features of the tumor. The microbiome is the population of microorganisms in a specific biological environment, and the gut microbiome has been implicated as one of the determinants for the efficacy of ICIs. The gut microbiome refers to the collective genomes of the microorganisms that reside in the digestive tract, and next-generation sequencing has allowed for culture-independent cataloging of entire genomes of individual organisms’ microbiomes. ^, The gut microbiome was first shown to be implicated in response to ICIs in patients with metastatic melanoma treated with ipilimumab, evaluated through fecal microbiota transfer to antibiotic-treated mouse models. Specific strains of Bacteroides ( B fragilis and/or B thetaiotaomicron ) and Burkholderiales cepacia in the gut were associated with improved anti-cancer responses. The first prospective study in humans evaluating the gut microbiota and its effect on response to ICIs was also done in metastatic melanoma, and found improved PFS and OS in patients whose microbiota was enriched with Faecalibacterium and other Firmicutes. This led to further research in melanoma, as well as in other tumors, including NSCLC. In a cohort of patients with NSCLC (n = 60) and renal cell carcinoma (n = 40), patients whose gut microbiome had greater numbers of microbial species (termed alpha-diversity), and the presence of Akkermansia mucinphila , there was an association with improved responses to ICIs. Interestingly, these data and others have observed that antibiotic use may be implicated in lack of response to ICIs, There is variation in specific bacterial taxa, which have demonstrated an association with response to ICIs in different studies, and this may potentially be due to different sequencing techniques. Importantly, there is currently no standardized approach to microbiome analysis. There is still much to be understood in the role of the microbiome in patients treated with ICIs, including continued identification of bacteria and metabolites as well as the actual mechanisms of the microbiome’s immunologic impact.

Neoadjuvant Immune Checkpoint Inhibitors in Non-small Cell Lung Cancer

Neoadjuvant ICI has shown promise in other solid tumors, such as melanoma and breast cancer. A mouse model of triple-negative breast cancer demonstrated improved survival in mice treated with neoadjuvant therapy compared with adjuvant therapy, and these mice also had an increase in tumor-specific T cells. A pilot study of neoadjuvant nivolumab for resectable NSCLC (n = 21) showed promising results, with no delays in surgery and a major pathologic response in 45% of patients. In addition, tumors with a major pathologic response had a higher frequency of T cell clones that had expanded in peripheral blood after PD-1 blockade. Neoadjuvant nivolumab is now being evaluated in a larger clinical trial. Of note, the availability of posttreatment resection specimens also allowed for in-depth histologic analysis, leading to a proposal for a standardized “Immune-Related Pathologic Response Criteria”, which was also demonstrated to be reproducible between pathologists. These criteria are now being used in other trials of neoadjuvant ICIs in NSCLC. The success with nivolumab sparked a neoadjuvant trial with atezolizumab for resectable NSCLC, with results of the initial safety analysis (n = 21) without major delays in surgery, and neoadjuvant atezolizumab was well tolerated. If results continue to be promising, it is likely that there will be a role for neoadjuvant ICI in the treatment of resectable NSCLC.