If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The advent of molecular targeted therapies and the more recent introduction of immune checkpoint inhibitors (ICIs) have altered and improved the therapeutic landscape of non-small cell lung cancer (NSCLC).
•
ICIs confer a durable response in a subset of patients; however, their therapeutic role in oncogene-driven NSCLC remains unclear, as the vast majority of trials was conducted without patients harbouring established oncogenic mutations.
•
The only randomised data available on the efficacy of ICIs on oncogene addicted NSCLC come from the IMpower 150 trial and concern only EGFR mutated NSCLC patients.
•
At present, no biomarker is clearly predictive for response or benefit to ICIs in oncogene addicted NSCLC. These are often characterised by low tumour mutation burden (TMB) and a less inflammatory tumour microenvironment, poor in tumour-infiltrating CD8+ lymphocytes. PD-L1 can be constitutionally overexpressed without having any clinical significance.
Introduction
Lung cancer is the leading cause of cancer-related death globally [
]. The majority of patients with NSCLC are diagnosed at advanced stages, where chemotherapy has only limited efficacy, at the price of significant toxicity [
Outcome of patients with a performance status of 2 in eastern cooperative oncology group study E1594: a phase III trial in patients with metastatic nonsmall cell lung carcinoma.
Can: Interdiscipl Int J Am Can Soc.2001; 92: 2639-2647
]. The advent of molecular targeted therapies against driver oncogenes such as EGFR mutations and ALK fusions have altered the therapeutic landscape of subsets of oncogene driven NSCLC [
] (Fig. 1). Despite these life-prolonging advances in NSCLC, the majority of patients ultimately acquire resistance to targeted therapies through a variety of mechanisms, resulting in cancer progression [
]. The recent introduction of immune checkpoint inhibitors (ICIs), including antibodies against programmed cell death-1 (PD-1), programmed cell death ligand-1 (PD-L1) and cytotoxic T-cell lymphocyte antigen-4 (CTLA4), can confer a durable response in a subset of patients, raising the hope of a “cure” [
Chemotherapy in combination with immune checkpoint inhibitors for the first-line treatment of patients with advanced non-small cell lung cancer: a systematic review and literature-based meta-analysis.
]. However, the therapeutic role of ICIs in oncogene-driven NSCLC remains unclear, as the vast majority of trials was conducted without patients harbouring established oncogenic mutations [
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial.
David F, Spigel R, Giaccone G, Reinmuth N, Vergnenegre A, Barrios C, et al., editors. IMpower 110: interimOS analysis of a phase III study of atezolizumab vs platinum-based chemotherapy as 1L treatment in PD-L1-selected NSCLC. 2019ESMO congress; 2019.
Whether ICIs have a role among these patients, and if so, when, is controversial. Most of the evidence currently available pertaining to immunotherapy in patients with oncogene-driven NSCLC comes from either subgroup analyses of clinical trials, small phase I or II non-controlled trials with combination regimens (generally including a targeted TKI as a backbone), or retrospective analyses from real-world data. In most ICI trials, patients with EGFR or ALK alterations are excluded, but the presence of other oncogenic drivers (e.g. MET, RET, ROS1, BRAF or KRAS) are neither detailed nor considered exclusion criteria. Consequently, the level of evidence supporting the use of immunotherapy in patients with NSCLC and driver mutations is quite low, and clinical decisions require a rigorous judgment balancing the benefit and harm to the patient.
The evolution of tumours bearing a molecular alteration usually depends on a single dominant mechanism following the principle of oncogenic addiction, which has been described as the dependence of tumour cells upon the specific activity of an activated oncogene [
]. A single mutation or translocation is supposed to confer a survival advantage to the respective cells, and it is usually isolated, explaining the low tumour mutation burden (TMB) observed in these tumours [
], and a less inflammatory tumour microenvironment, poor in tumour-infiltrating CD8+ lymphocytes. These facts might, therefore play a role in the lack of sensitivity to ICIs in oncogene-driven lung cancer [
Herein, we explore the role and impact of ICIs in NSCLC harbouring other oncogenic driver alterations and discuss the possible biological rationale to explain the lack of sensitivity to ICIs in these diseases
EGFRxxx
The monomeric transmembrane epidermal growth factor receptor, EGFR, is a receptor tyrosine kinase involved in major cell proliferation pathways. It primarily exerts this function through the RAS-RAF-MAPK-MEK, PI3K-AKT-mTOR and STAT pathways. EGFR activating mutations are among the most frequent oncogenic alterations in NSCLC. In lung cancer, they are most prevalent among never or light smokers and young, Asian, female patients. The incidence of EGFR mutations varies significantly by ethnicity. Among Caucasians with any stage of lung adenocarcinoma, they occur in 15% to 27% of patients, depending on the stage, while this can reach 62% in East Asians with advanced disease [
EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.
Clin Can Res: Off J Am Assoc Can Res.2016; 22: 4585-4593
The efficacy of ICIs among patients with EGFR activating mutations in NSCLC appears to be quite limited, though most data are from early phase or retrospective trials [
]. In the phase I CA209-012 and Keynote 001 trials, front-line ICIs provided lower response rates (14% versus 30% and 16% versus 37%, respectively) in patients with EGFR mutant versus wild-type advanced NSCLC [
]. The poor response was independent of PD-L1 status. Similarly, survival results were inferior in the EGFR subgroups.
In a phase II trial of upfront pembrolizumab in TKI-naïve EGFR mutated lung cancer patients, the RR was 0% after the first 11 patients, leading to early trial discontinuation [
The phase II BIRCH trial, assessing the PD-L1 inhibitor, atezolizumab, in different lines of therapy in NSCLC, included a subgroup analysis about EGFR mutant patients. In first-line, the response was similar among the 13 EGFR mutant patients (19%) was similar to that among wild-type (23%) patients, though the small numbers makes interpreting these results difficult. Among the 32 remaining EGFR mutant patients in second-line or above, there was almost no response and outcomes were far inferior to those of wild-type patients [
Phase II trial of atezolizumab as first-line or subsequent therapy for patients with programmed death-ligand 1-selected advanced non-small-cell lung cancer (BIRCH).
J Clin Oncol: Off J Am Soc Clin Oncol.2017; 35: 2781-2789
In the phase II, single-arm, ATLANTIC trial durvalumab was assessed among pretreated NSCLC patients. Among EGFR or ALK altered patients, outcomes were inferior to wild-type patients, independently of PD-L1 expression [
]. The only randomised data available on the efficacy of ICIs in EGFR mutated NSCLC patients come from the IMpower 150 trial, a randomised phase 3 study in first-line treatment with patients randomised (1:1:1) to receive atezolizumab-bevacizumab-carboplatin-paclitaxel (ABCP), atezolizumab-carboplatin-paclitaxel (ACP), or bevacizumab carboplatin-paclitaxel BCP every three weeks [
Atezolizumab plus bevacizumab and chemotherapy in non-small-cell lung cancer (IMpower150): key subgroup analyses of patients with EGFR mutations or baseline liver metastases in a randomised, open-label phase 3 trial.
]. Efficacy was assessed in key subgroups within the intention-to-treat population, including patients with EGFR mutations (both sensitising and non-sensitising; EGFR-positive) previously treated with one or more tyrosine kinase inhibitors. The analysis showed an improved OS with ABCP versus BCP in patients with sensitising EGFR mutations (median overall survival not reached with ABCP [26/400] vs 17·5 months with BCP [32/400]; HR 0·31 [95% CI 0·11–0·83]). Despite its limitations, IMpower 150 remains the only study that included and clearly analysed data on an EGFR mutated population treated with a combination of chemotherapy and ICIs.
ALKxxx
ALK is a transmembrane receptor tyrosine kinase that consists of an extracellular ligand-binding domain, a transmembrane domain, and an intracellular tyrosine kinase domain [
], several ALK gene alterations have been identified across different tumour types, including point mutations, deletions, insertions and rearrangements leading to ALK reactivation. The Echinoderm Microtubule-associated protein-Like 4 (EML4)-ALK is the most prevalent in NSCLC [
Multicenter immunohistochemical ALK-testing of non–small-cell lung cancer shows high concordance after harmonization of techniques and interpretation criteria.
]. When tumour growth is driven by constitutive activation of the ALK fusion oncogene, they are susceptible to an ATP analogue inhibitor of ALK. There are actually 5 ALK tyrosine kinase inhibitors (TKIs) that have shown superior efficacy and a more tolerable toxicity profile than chemotherapy in first-line in patients with advanced NSCLC and an ALK [
The data about ICI efficacy in patients with NSCLC harbouring an ALK rearrangement are scarce but suggests very poor activity of single agent ICIs. In the phase 2 ATLANTIC trial, the efficacy of durvalumab was evaluated in NSCLC patients who had received at least two prior lines systemic therapy. Fifteen patients harboured ALK alterations, and no response was registered in that group [
]. Similar data were published on the IMMUNOTARGET study where 23 ALK-rearranged patients were treated with ICIs, without any responses. Further supporting these results, the Massachusetts General Hospital oncologist group published a retrospective analysis on 6 patients with ALK-rearranged NSCLC treated with ICIs, showing again no response [
EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.
Clin Can Res: Off J Am Assoc Can Res.2016; 22: 4585-4593
]. Further data were presented in a retrospective multicenter French study in which 2 out of 8 ALK-positive patients, treated with ICIs, achieved a partial response [
Efficacy and safety of programmed cell-Death-Protein-1 and its ligand inhibitors in pretreated patients with epidermal growth-Factor receptor-mutated or anaplastic lymphoma kinase-Translocated lung adenocarcinoma.
Finally, an arm of the phase 3 IMpower150 trial evaluated the safety and efficacy of chemotherapy with atezolizumab and bevacizumab in non-squamous metastatic NSCLC [
]. Thirteen percent of patients harboured ALK rearrangements or EGFR mutation, all pre-treated with TKIs. For the analysis, ALK-positive patients were grouped together with patients harbouring an EGFR mutation, making any interpretation quite difficult. In the 34 ALK-positive patients included in the study, no statistically significant differences were observed in PFS for the quadruplet combination compared with the bevacizumab/chemotherapy arm (8.3 vs 5.9 months; HR, 0.65; nonsignificant) [
]. Similar findings were seen in the IMpower130 trial, in which 44 patients with NSCLC had ALK-rearranged tumours, and the EGFR/ALK group obtained similar outcomes in both treatment groups of chemotherapy with and without atezolizumab [
Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicentre, randomised, open-label, phase 3 trial.
Though these data are scant, ICIs as single agents do not appear promising in ALK-rearranged NSCLC. Chemotherapy remains the standard of care after exhaustion of ALK TKIs. Whether to add ICIs to chemotherapy or even consider a quadruplet, as per IMpower 150, remains unclear and further prospective data are warranted to address the question.
ROS1xxx
The ROS1 gene belongs to the subfamily of tyrosine kinase insulin receptor genes [
]. ROS1 fusions are identified in about in 2.5% of the patients with lung adenocarcinoma and at even higher frequencies in spitzoid neoplasms and inflammatory myofibroblastic tumours [
]. So far, 26 genes have been identified as fusion partners for ROS1. Interestingly, some of them can fuse with RET and ALK. All the fusion proteins retain the ROS1 kinase domain, but rarely its transmembrane domain. Most of the partners have dimerization domains that are retained in the fusion, presumably leading to constitutive ROS1 tyrosine kinase activation. Some partners have transmembrane domains that are retained or not in the chimeric proteins. Several drugs that bind to the ATP-binding site of their target enzymes are now used or under development. There are several tyrosine kinase inhibitors approved by the Food and Drug Administration (FDA) for the treatment of ROS1 positive NSCLC: crizotinib, lorlatinib, entrectinib [
]. The efficacy of ICIs in patients with ROS1 fusion-positive metastatic NSCLC patients is unclear. However, the IMMUNOTARGET registry offers a pessimistic outlook, with 5 of the 6 included patients presenting with progressive disease to single-agent ICIs, and a response rate of 17% (1 patient), with a median OS of 18.4 months [7.0; NR [
A case report about a 52-year-old male former (15 pack-years) smoker, diagnosed with metastatic lung adenocarcinoma, was recently published. He received first-line treatment with carboplatin-pemetrexed and bevacizumab with a good response, followed by maintenance with pemetrexed and bevacizumab for more than one year. Due to progression, the patient was started on nivolumab (anti-PD-1) achieving partial response. Seven months thereafter, the original lung biopsy specimen was sent for screening as part of a clinical trial, and a ROS1 gene rearrangement was identified. Given the excellent response, nivolumab was continued and a scan performed a few months later showed a complete radiographic response, sustained for more than two years at the time of the publication. This patient was found to have a novel ROS1 fusion with non-muscle heavy chain 9 gene (MYH9) [
]. The breakpoint occurring in ROS1 exon 36 had not been previously reported. Both wild-type and fusion ROS1 transcription were found in this patient. It was the first report of a ROS1-rearranged lung adenocarcinoma demonstrating a complete radiographic response to ICIs, with a progression-free survival of 3.5 years. This unique case highlights the potential role of ICIs in ROS1 fusion-positive NSCLC. From these anecdotal cases, it may be possible that at least a subset of patients with ROS1 rearrangements may be immunogenic, with the sensitivity to ICIs based on the interaction between tumour microenvironment and immune specific factors and not simply the interaction between the PD-1 and PD-L1 axes.[
] Furthermore, it is possible that MYH9-ROS1 may not be associated with constitutive kinase activation, explaining the efficacy of immunotherapy in this context. However, given the lack of data, and effective alternatives, we do not consider ICIs to be standard front-line therapy in NSCLC with ROS1 rearrangements and would reserve this option for late treatment lines, after targeted agents.
BRAF p.V600Exxx
BRAF p.V600E mutation are seen in 2% of NSCLC. BRAF inhibitors either as monotherapy or in combination have shown substantial anti-tumour activity in both treatment naïve and treatment-refractory NSCLC harbouring BRAF p.V600E mutation [
Efficacy of vemurafenib in patients with non–small-cell lung cancer with BRAF V600 mutation: an open-label, single-arm cohort of the histology-independent VE-BASKET study.
]. Dabrafenib and trametinib are currently FDA approved for BRAF p.V600E alteration. Several MAPK and ERK inhibitors are in development for non-BRAF V600 alterations. Conflicting data surround the role of ICIs among NSCLC patients harbouring BRAF V600 mutations [
]. In the first, of 21 evaluable patients treated with ICIs, 12 had the most common BRAF variant, p.V600E, and 9, non-p. V600E mutations. The median PFS was of 3.7 and 4.1 months among patients with p.V600E and non-p.V600E variants, respectively. At the time of publication, the median OS was not reached. The objective response rate (ORR) was 25% among patients with p.V600E mutations and 33% among the others. These results mirrored those reported in the second retrospective analysis, this time on 38 patients with BRAF mutations. It found an ORR of 28%, a median PFS of 3.0 months, and an OS of 13.1 months.
A smaller retrospective study reported the efficacy of ICIs among 3 NSCLC patients with BRAF mutations. These patients exhibited a numerically inferior OS compared to similar patients receiving front-line chemotherapy [
]. While front-line ICIs in BRAF mutant NSCLC do not appear to be very promising, their efficacy in second or greater line seems to be comparable to that of ICIs in non-oncogene driven NSCLC [
Guisier F, Dubos-Arvis C, Viñas F, Doubre H, Ricordel C, Ropert S, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced Non Small Cell Lung Cancer with BRAF, HER2 or MET mutation or RET-translocation. GFPC 01-2018. J thorac oncol. 2020.
]. Among 44 BRAF-mutated NSCLC patients, comprising 26 with p.V600E variants and 18 with others subtypes, the ORR was roughly 30%, with 26% ORR for p.V600E variants and 33% for other variants. While patients’ numbers are small and data are not prospective, these results nonetheless suggest that patients with NSCLC and BRAF mutations exhibit some degree of sensitivity to ICIs. BRAF mutations, more commonly found among smokers in NSCLC, could more closely mirror the impact and behaviour of KRAS mutations than those of EGFR or ALK alterations. From a biological point of view, given BRAF’s association with smoking, PD-L1 expression and a higher mutational burden, there is a rationale for a higher sensitivity to ICIs compared to some other oncogenic alterations [
Chemotherapy in combination with immune checkpoint inhibitors for the first-line treatment of patients with advanced non-small cell lung cancer: a systematic review and literature-based meta-analysis.
First-in-class ERK1/2 inhibitor ulixertinib (BVD-523) in patients with MAPK mutant advanced solid tumors: results of a phase I dose-escalation and expansion study.
In clinical practice, we favour front-line targeted therapy for BRAF p.V600E mutation but if this is unavailable, we would use chemo-immunotherapy among these patients.
NRG1xxx
Neuregulin 1 (NRG1) fusions are detected in roughly 0.3% of NSCLCs [
]. In a recent retrospective analysis of 117 patients with NSCLC and NRG1 fusions, 95% of cases were among adenocarcinoma, most frequently of the mucinous subtype, almost half of the patients were non-smokers and there was slight over-representation among female patients. Anecdotal and retrospective of NRG1 gene fusions positive NSCLC treated with HER2/HER3 inhibitors reveal the sensitivity of this fusion to agents like afatinib [
]. Several other agents are in clinical development. Among 18 responsive-evaluable patients treated with a platinum-doublet chemotherapy, there was an 11% ORR and 61% disease control rate. On the other hand, among the 6 patients treated with ICIs alone and 5 patients treated with chemo-immunotherapy, no responses were observed [
] do not support routine use of ICIs, especially if alternatives are available. Today, we would reserve ICIs for late treatment lines.
NTRKxxx
NTRK gene fusions have the peculiarity of occurring in adenocarcinoma, squamous and even neuroendocrine lung cancers. In NSCLC, it can be detected in roughly 0.2% of cases. No particular clinical characteristics have been identified, namely there does not appear to be a correlation with sex, age, nor smoking status [
]. These fusions are true oncogenic drivers and are mutually exclusive with other drivers. The FDA has approved Larotrectinib and entrectinib for NTRK fusion-positive tumors. In NSCLC, NTRK gene fusions are associated with higher TMB and PD-L1 expression than EGFR, ALK and ROS1 alterations [
]. Unfortunately, there are currently no data about the efficacy of ICIs among these patients. In practice, if an NTRK inhibitor is available, we would favour this option, and if it is not, we would give patients combined chemo-immunotherapy to avoid potentially undertreating patients.
METxxx
MET is a high-affinity proto-oncogene receptor tyrosine kinase playing an important role in cell proliferation, survival, and metastases [
]. Select somatic alterations in MET lead to an alternatively spliced transcript that is a result of exon 14 skipping occurring in 3%–5% of lung cancers, while de novo amplifications are found in 1–5% of NSCLC patients [
]. Crizotinib, capmatinib, savolitinib and tepotinib all have shown activity in this subset. Capmatinib was recently FDA approved for this mutation.
To date, the response to ICIs has not been well characterized. Recently, a retrospective study has investigated the outcomes of patients with MET exon 14-altered lung cancers treated with ICIs, showing an ORR of 17% (4/24) and median PFS of 1.9 months (95% CI 1.7–2.7). Although some responses to ICIs were reported, the overall clinical efficacy of ICIs was quite modest [
]. Patients with MET alterations (n = 36) were included, of which 23 with exon 14 skipping and 13 with amplifications. The ORR was 17%, the median PFS was 3.4 [1.7; 6.2] months, and the MET exon 14 cohort showed a numerically better PFS of 4.7 months, though this was not statistically significant [
]. Neither PD-L1 nor TMB was associated with outcomes. One noteworthy result was that 23.4% of patients with MET alterations were long-term responders to ICIs, second only to KRAS mutated NSCLC, suggesting a subset of checkpoint-sensitive patients. This result was supported by a recent limited series (n = 8) from Dudnik et al. in which the median PFS was 4.0 months (95% CI, 2.4–not reached) [
]. A third retrospective study mirrored these results. Among 30 patients with NSCLC harbouring MET mutations and treated with ICIs, there was a 35.7% ORR and 4.9-month PFS [
]. All three studies had a limited number of patients making the results difficult to interpret and failing to clarify the possible limits and benefit of using ICIs in this context.
If targeted MET therapy is available, we favour its use in the first-line, as using it after ICIs could increase the former’s toxicity. In this case, we recommend ICIs in subsequent lines, preferring the combination with chemotherapy.
HER2xxx
HER2 alterations in NSCLC include HER2 mutations and amplifications. These are each detected in approximately 2–5% of adenocarcinomas. HER2 pathway monoclonal antibody drug conjugates like ado-trastuzumab emtansine (T-DM1), fam-trastuzumab deruxtecan-nxki (T-DXd, also known as DS8201a) and TKIs such as neritinib, afatinib, poziotinib, pyrotinib, lapatinib have shown activity in this subset. T-DXd has received FDA breakthrough therapy designation for HER2 mutant NSCLC. PD-L1 expression among HER2 mutant NSCLCs is lower than in unselected lung cancers, but TMB values are similar [
]. There is no prospective evidence about the efficacy of ICIs for patients with lung cancer harbouring HER2 alterations. In a retrospective study from the MSKCC, 26 NSCLC patients harbouring HER2 mutations, treated with ICIs, had an ORR of 12%, with a median duration of response of about 3.4 months, PFS of 1.9 months and OS of 10.4 months [
] there were 29 patients harbouring HER2 mutations. Here, there was also a low ORR, at 7.4%, with a median PFS of 2.5 months and OS of 21.3 months. Another retrospective study, this time comprising 23 patients with HER2-mutant NSCLC, found a 27.3% ORR to ICIs. The PFS and OS were 2.2 and 20.4 months respectively [
Guisier F, Dubos-Arvis C, Viñas F, Doubre H, Ricordel C, Ropert S, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced Non Small Cell Lung Cancer with BRAF, HER2 or MET mutation or RET-translocation. GFPC 01-2018. J thorac oncol. 2020.
]. The heterogeneity of responses suggests some subtypes of HER2-mutant NSCLC may be sensitive to ICIs, and the results are similar to unselected populations in later treatment lines. Here, in the absence of clinical trials, we would use chemo-immunotherapy as a standard front-line approach.
KRAS p.G12Cxxx
KRAS mutations represent the most frequent molecular alterations encountered in advanced NSCLC. KRAS mutations are heterogeneous and can result in substitutions involving codons 12, 13, or 61[
]. The most frequent substitution, found in 41% of KRAS-mutant NSCLC, is KRAS p.G12C. This mutation is commonly identified in patients with a substantial history of smoking and recently, for the first time a p.G12C tyrosine kinase inhibitor has shown some efficacy [
]. In contrast, KRAS p.G12D substitutions are more commonly found in tumours of patients with little to no prior history of smoking. Sotorasib (AMG 510) and adagrasib (MRTX849) covalent KRAS G12 C inhibitors are in clinical trials and have demonstrated early activity in this subset of NSCLC.
KRAS-mutant NSCLC generally exhibits increased tumour mutation load, potentially leading to increased ICI sensitivity [
], unlike other driver mutations which tend to exhibit a cold immune-environment. This might stem from the increased incidence of select KRAS mutations in smokers, as smoking is associated with elevated somatic DNA mutations found in tumour cells [
]. Furthermore, concomitant alterations may affect the immunogenicity of KRAS-mutant tumours. Co-occurring TP53 mutations are often associated with enhanced tumour cell proliferation and inflammation, allowing for an immune-rich micro-environment. On the other hand, co-occurring STK11 mutations, present in roughly 20% of KRAS-mutant NSCLC, can reduce immune surveillance, possibly by modulating the NF-kB pathway. Some data suggest that concurrent KRAS and STK11 mutations in NSCLC patients may be predictive of primary resistance to ICIs [
Clinical outcome of ALK-positive non–small cell lung cancer (NSCLC) patients with De Novo EGFR or KRAS co-mutations receiving tyrosine kinase inhibitors (TKIs).
]. In contrast, co-occurring TP53 mutations seem to be associated with enhanced tumour cell proliferation and inflammation, allowing for an immune-rich micro-environment.. Cancers with concomitant KRAS-mutant and STK11-mutant are associated with decreased TILs, contributing to the suppression of immune surveillance [
Differential association of STK11 and TP53 with KRAS mutation-associated gene expression, proliferation and immune surveillance in lung adenocarcinoma.
Given these results from retrospective data, and conflicting results in a post-hoc analysis of Keynote 042, there is no consensus as to the current clinical implications of these mutations [
Abstract CT084: Relationship between STK11 and KEAP1 mutational status and efficacy in KEYNOTE-042: pembrolizumab monotherapy versus platinum-based chemotherapy as first-line therapy for PD-L1-positive advanced NSCLC.
]. In a meta-analysis exploring the impact of KRAS mutations on the efficacy of ICIs in the second-line setting, the presence of a KRAS mutation appears predictive of superior outcomes of ICIs compared to docetaxel. At the same time, there is no difference between treatments in the KRAS wild-type population [
]. Similarly, in an exploratory analysis based on the Keynote 042 trial of front-line pembrolizumab in advanced NSCLC, the role of KRAS mutations appears to be predictive of numerically superior ORR (57 vs 29%), median PFS (12 vs 6 months) and OS (28 vs 15 months). The difference is even more pronounced in the KRAS p.G12C subset, though all of these investigations are exploratory [
LBA4 Association of KRAS mutational status with response to pembrolizumab monotherapy given as first-line therapy for PD-L1-positive advanced non-squamous NSCLC in KEYNOTE-042.
]. Given their proven efficacy, we readily use ICIs in patients with NSCLC harbouring KRAS mutations, either as single-agent for high PD-L1 expressing tumours, or in combination with chemotherapy or other immunotherapy. We opt for targeted therapy alone in the context of clinical trials, when and if available.
RETxxx
RET (rearrangement during transfection) is a tyrosine kinase receptor that, when fused with a partner molecule, activates oncogenic activity and promotes unchecked cellular proliferation [
]. It has been demonstrated that pemetrexed-based chemotherapy and multikinase inhibitors have modest effect for patients with RET-rearranged lung cancers [
],2,3 Recently, highly selective RET inhibitors, selpercatinib and pralsetinib have been FDA approved for RET fusion-positive NSCLC. Given the impressive response rates with TKIs [
Moretti SBVLR, Bellezza G, Metro ASFRG. RET Rearrangement as a Predictor of Unresponsiveness to Immunotherapy in Non-Small Cell Lung Cancer: Report of Two Cases with Review of the Literature.
]. Currently, four retrospective studies have assessed the efficacy of ICIs in patients with RET-rearranged lung cancer. In a retrospective study that included 551 patients, 16 patients had RET-rearranged NSCLC [
]. The majority was diagnosed with adenocarcinoma and treated with the PD-1 inhibitors nivolumab or pembrolizumab. Approximately 66.7% of patients with RET-rearranged NSCLC were never-smokers, 26.7% were former smokers, and 6.7% were current smokers. The median number of cells positive for PD-L1 expression was 26%. Patients were followed for a median of 16.1 months, and the ORR among patients with RET rearrangements was 6%, and progressive disease was observed in 75% of patients. The median OS was 21.3 months (95% confidence interval [CI], 3.8–28.0), while the median PFS was 2.1 months (95% CI, 1.3–4.7). The modest results suggest that immune checkpoint inhibitors should not be used as single agents in patients with RET-rearranged NSCLC.
A retrospective, multicenter study included 107 patients who were mostly treated with the PD-1 inhibitors nivolumab and pembrolizumab.6 Approximately 56% had PD-L1 expression ≥ 50%. The study specifically included nine patients with NSCLC with RET rearrangements, of whom the majority had adenocarcinoma. Overall, 44% of patients were never-smokers, 33% former smokers, and 22% active smokers. Patients were followed for a median of 9.2 months. Patients with RET rearrangements had an ORR of 37.5%, the median duration of response was 12.1 months (95% CI, 8.4-not reached [NR]), median PFS was 7.6 months (95% CI, 2.3-NR), and median OS was not reached. It is important to note that the higher response rates in this study may be related to the fact that this patient population received immunotherapy as an earlier line of treatment than in other studies [
Guisier F, Dubos-Arvis C, Viñas F, Doubre H, Ricordel C, Ropert S, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced Non Small Cell Lung Cancer with BRAF, HER2 or MET mutation or RET-translocation. GFPC 01-2018. J thorac oncol. 2020.
]. Another study of 74 patients with RET-rearranged included NSCLC patients who were treated with pembrolizumab, nivolumab, atezolizumab, durvalumab, or ipilimumab plus nivolumab.2 Approximately 69% of patients were never-smokers and 31% were former smokers. PD-L1 expression ranged from 0% to ≥50% (58% of patients had 0% PD-L1 expression; 23% had 1%-49% PD-L1 expression; and 19% had ≥50% PD-L1 expression).2 Among 13 patients assessed for response, disease progression was observed in 62% of patients. The PFS was 3.4 months (95% CI, 2.1–5.6). One patient in this cohort had high PD-L1 expression (50%), and despite this, responded poorly to dual immune checkpoint inhibitor therapy. These findings are consistent with previous studies showing that immune checkpoint inhibitors have little effect in patients with RET-rearranged NSCLC. Finally, in a similar retrospective study of 59 patients with RET rearrangements in South Korea, approximately 51% of the total cohort comprised never-smokers, 28.8% were former smokers, 20.3% were current smokers, and the majority had adenocarcinoma [
Guisier F, Dubos-Arvis C, Viñas F, Doubre H, Ricordel C, Ropert S, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced Non Small Cell Lung Cancer with BRAF, HER2 or MET mutation or RET-translocation. GFPC 01-2018. J thorac oncol. 2020.
]. In this cohort, 13 patients were treated with ICIs. The ORR in this patient population was 7.7%, the median PFS was 2.1 months (95% CI, 1.6–2.6), and the median OS was 12.4 months (95% CI, 2.9–21.8). Overall, patients in this study with RET rearrangements did not experience clinical benefit from ICIs [
]. In a single institutional study from the MD Anderson Cancer Center, among 70 patients who received systemic therapy for RET + malignancies, non-ICI therapy was associated with decreased risk for treatment discontinuation compared with ICI in the overall population (HR = 0.31; 95% CI 0.16–0.62; p = 0.000834 [
In summary, while we lack prospective data, current evidence suggests that RET-rearranged NSCLCs, which are biologically cold tumours, are not very sensitive to ICIs and that, when available, targeted therapies should be preferred [
TP53 represents one of the most studied genes in humans. It is involved in cell-cycle chepoint regulation, stimulation of DNA damage repair and induction of apoptosis [
]. TP53 mutations are independently correlated with longer OS in patients with advanced NSCLC. A possible explanation stems from the connection between TP53 and TMB. Dysfunctional TP53 could result in an accumulation of mutations, generating potentially immunogenic neoantigens and a high TMB. The presence of TP53 mutations also appears to increase the expression of immune-checkpoint receptors and an activated T-effector and interferon-signature. It is unclear how co-mutation of TP53 with oncogenes or other tumor suppressor genes influences the response to checkpoint inhibitors.
] suggest that co-mutation of TP53 with KMT2C, another oncogene often mutated in NSCLC, confer a favorable response to ICIs. In their cohort, co-occurring KMT2C mutations significantly enhanced the response of NSCLC patients to ICIs, serving as proof of principle that finer patient stratification is more informative to guide clinical decision. In the same study, the authors also considered the role and possible impact other two tumour suppressor genes, STK11 and KEAP1. In this constellation, they did not significantly alter the response of NSCLC patients to ICIs. As gene function depends on context and co-occuring mutations, any potential influence of these genes in NSCLC is not ruled out.
Biology: Why are oncogene driven tumours poorly responsive to immunotherapy?
Role of TMB
Lung cancer in never or minimal smokers is generally associated with a low TMB, which results in a lack of immunogenic neo-antigens, and thus a non-inflamed (“excluded”) microenvironment [
EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.
Clin Can Res: Off J Am Assoc Can Res.2016; 22: 4585-4593
]. This is a possible explanation for the more favourable outcomes observed in the KRAS, BRAF non-V600E, and even MET exon14 altered NSCLC patients, as these oncogenes are more frequently observed in smokers. These findings are in line with other studies reporting low TMB in EGFR, ALK, ROS1 [
]. NSCLC harbouring EGFR mutations are often associated with low TMB; however, recent data have highlighted that high TMB could have a potential negative prognostic role in EGFR mutant NSCLC treated with EGFR TKIs [
]. Furthermore, a recent publication by Hastings et al showed that EGFR p.L858R and G719 (exon 21 and 18) tumours have higher TMB compared to EGFR dels19 tumours, in line with slightly worse outcomes with ICIs in patients harbouring EGFR dels19 mutant NSCLC compared to those with wild-type EGFR [
]. These could impact and alter the tumour microenvironment, leading to immunogenic cell death with neoantigen release and local and systemic T cell expansion. The change in the microenvironment along with DNA alterations, with concomitant increases of TMB, could explain the sensitivity to ICIs seen in some never-smokers with NSCLC. It is likely that these changes, combined with DNA alterations (and therefore increased TMB), explain the benefit observed in never-smokers. TMB analysis may, therefore, be a useful biomarker in this population (Fig. 2)
Fig. 2Cold tumours (left) have an immune-poor microenvironment, rich in regulatory T cells but poor in activated T cells. These are associated with low tumour mutation burden and PD-L1 expression. Hot tumours (right) have an immune-rich microenvironment with activated T cells. These are associated with high TMB and PD-L1 expression and respond better to checkpoint inhibitors
PD-L1 level expression is not a reliable and robust predictor of response to ICIs, even when very elevated, in patients with oncogene addicted NSCLC. Four main mechanisms may explain this finding: (1) alterations of genes, including EGFR, ALK fusions, KRAS, MYC, PTEN, and p53; (2) high levels of exogenous inflammatory cytokines, such as interferon-γ; (3) PD-L1 amplification [
Chemotherapy in combination with immune checkpoint inhibitors for the first-line treatment of patients with advanced non-small cell lung cancer: a systematic review and literature-based meta-analysis.
] disruption of the 3 j-untranslated region of the PD-L1 gene. Furthermore, in such tumours, there is rarely a strong immune cell infiltration, which is essential to stimulate an immune-response [
]. There have been reports investigating the relationship between PD-L1 and oncogene-driven NSCLC, in particular EGFR, ALK and KRAS that we have summarized in Table 2.
Table 1Oncogenic Drivers and Response to Immunotherapy.
EGFR Mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.
Clin Can Res An Off J Am Assoc Can Res.2016; 22: 4585-4593
Efficacy and safety of programmed cell-Death-Protein-1 and its ligand inhibitors in pretreated patients with epidermal growth-Factor receptor-mutated or anaplastic lymphoma kinase-Translocated lung adenocarcinoma.
Guisier F, Dubos-Arvis C, Viñas F, Doubre H, Ricordel C, Ropert S, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced Non Small Cell Lung Cancer with BRAF, HER2 or MET mutation or RET-translocation. GFPC 01-2018. J thorac oncol. 2020.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial.
Programmed cell death-ligand 1 expression in surgically resected stage I pulmonary adenocarcinoma and its correlation with driver mutations and clinical outcomes.
Clinicopathologic analysis of programmed cell death-1 and programmed cell death-ligand 1 and 2 expressions in pulmonary adenocarcinoma: comparison with histology and driver oncogenic alteration status.
Tumour microenvironment and tumour infiltrating lymphocytes
The tumour microenvironment (TME) plays a crucial role in immunogenicity, in particular of oncogene-driven lung cancer. The interaction between cancer and immune cells may result in a proliferation of regulatory T cells, downplaying and downregulating tumour infiltrating lymphocytes and major histocompatibility complex [
Role of the dynamic tumor microenvironment in controversies regarding immune checkpoint inhibitors for the treatment of non-small cell lung cancer (NSCLC) with EGFR mutations.
]. Furthermore, the density and diversity of tumour infiltrating immune cells could be related to prognosis and prediction of treatment efficacy. Two major subsets of tumours have been identified so far. They are characterized by the presence/absence of gene expression profile modulating the T-cell–inflamed tumour microenvironment. The T-cell–inflamed subset of tumours, also called “hot tumours”, contains T-cell transcripts, chemokines that mediate effector T-cell recruitment, macrophage activation, and a type I interferon (IFN) transcriptional profile. These tumours are characterised by high infiltration of CD8+ T cells, macrophages, some B cells and plasma cells [
Differential association of STK11 and TP53 with KRAS mutation-associated gene expression, proliferation and immune surveillance in lung adenocarcinoma.
]. This inflammatory signalling results in the attraction of various immune cellular populations, including tumour-associated macrophages (TAMs), tumour-reactive lymphocytes, myeloid-derived suppressor cells (MDSCs), tumour-associated neutrophils and mast cells. These interact with tumour cells to ultimately shape a highly immune suppressive TME, with diminished tumour cytotoxic and enhanced tumour-promoting manifestations [
]. MDSCs are immune suppressive myeloid cells fostering tumour progression in many different ways, most of which result in the inhibition of activation of tumour-reactive T cells and of natural killer (NK) cell cytotoxicity [
]. On the contrary “not inflamed” or “cold tumours”, have low immune cell infiltration and have been described to be immunologically ignorant. They are characterised by low TMB and low expression of antigen presentation machinery markers such as major histocompatibility complex class I (MHC I). Notably, it has been documented that oncogene activation downregulates expression of MHC I at the cell surface, thus impairing recognition by CD8 + T cells and promoting immune evasion [
]. Oncogene-addicted lung cancers mostly exhibit a phenotype resembling cold tumours, with striking features of T cell absence or exclusion within the TME and poor response to ICIs.
Conclusion
Taking into consideration the currently available body of literature, it is important to highlight that all oncogene-driven tumours are not equal and certainly not a homogenous group. It is tempting to try to group all of these drivers as a single entity, but neither patient nor tumour characteristics are uniform, both within groups of alterations like KRAS, MET or BRAF and between oncogenic drivers. As precision medicine gains ground, this important heterogeneity will require further research and analysis. Perhaps the tumour micro-environment and immunogenicity will be a key to better understanding the biology of each cancer and personalising the therapeutic approach. Today, in NSCLC patients with the exception of KRAS mutations likely related to smoking status, the established oncogenic drivers appear less sensitive to ICIs than unselected tumours (Table 1). Given the impressive results from selective TKIs and the risk of enhanced toxicity from administering certain TKIs after ICIs, we favour the approach of front-line TKIs, when available. When this is not an option, we would recommend a chemo-immunotherapy combination rather than ICIs alone irrespective of the PD-L1 expression, given the generally poor performance of the latter in oncogene-driven NSCLC. Combinations of targeted therapy and ICI, available in the context of clinical trials, may be more appealing than ICI monotherapy, though data are currently unavailable. Any future trial of ICIs in NSCLC should consider oncogenic drivers beyond EGFR and ALK, to help real-life clinical decision making.
References
Molina J.R.
Yang P.
Cassivi S.D.
Schild S.E.
Adjei A.A.
Non-small cell lung cancer: epidemiology, risk factors, treatment, and survivorship.
Outcome of patients with a performance status of 2 in eastern cooperative oncology group study E1594: a phase III trial in patients with metastatic nonsmall cell lung carcinoma.
Can: Interdiscipl Int J Am Can Soc.2001; 92: 2639-2647
Chemotherapy in combination with immune checkpoint inhibitors for the first-line treatment of patients with advanced non-small cell lung cancer: a systematic review and literature-based meta-analysis.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial.
David F, Spigel R, Giaccone G, Reinmuth N, Vergnenegre A, Barrios C, et al., editors. IMpower 110: interimOS analysis of a phase III study of atezolizumab vs platinum-based chemotherapy as 1L treatment in PD-L1-selected NSCLC. 2019ESMO congress; 2019.
EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.
Clin Can Res: Off J Am Assoc Can Res.2016; 22: 4585-4593
Phase II trial of atezolizumab as first-line or subsequent therapy for patients with programmed death-ligand 1-selected advanced non-small-cell lung cancer (BIRCH).
J Clin Oncol: Off J Am Soc Clin Oncol.2017; 35: 2781-2789
Atezolizumab plus bevacizumab and chemotherapy in non-small-cell lung cancer (IMpower150): key subgroup analyses of patients with EGFR mutations or baseline liver metastases in a randomised, open-label phase 3 trial.
Multicenter immunohistochemical ALK-testing of non–small-cell lung cancer shows high concordance after harmonization of techniques and interpretation criteria.
Efficacy and safety of programmed cell-Death-Protein-1 and its ligand inhibitors in pretreated patients with epidermal growth-Factor receptor-mutated or anaplastic lymphoma kinase-Translocated lung adenocarcinoma.
Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicentre, randomised, open-label, phase 3 trial.
Efficacy of vemurafenib in patients with non–small-cell lung cancer with BRAF V600 mutation: an open-label, single-arm cohort of the histology-independent VE-BASKET study.
Guisier F, Dubos-Arvis C, Viñas F, Doubre H, Ricordel C, Ropert S, et al. Efficacy and safety of anti-PD-1 immunotherapy in patients with advanced Non Small Cell Lung Cancer with BRAF, HER2 or MET mutation or RET-translocation. GFPC 01-2018. J thorac oncol. 2020.
First-in-class ERK1/2 inhibitor ulixertinib (BVD-523) in patients with MAPK mutant advanced solid tumors: results of a phase I dose-escalation and expansion study.
Clinical outcome of ALK-positive non–small cell lung cancer (NSCLC) patients with De Novo EGFR or KRAS co-mutations receiving tyrosine kinase inhibitors (TKIs).
Differential association of STK11 and TP53 with KRAS mutation-associated gene expression, proliferation and immune surveillance in lung adenocarcinoma.
Abstract CT084: Relationship between STK11 and KEAP1 mutational status and efficacy in KEYNOTE-042: pembrolizumab monotherapy versus platinum-based chemotherapy as first-line therapy for PD-L1-positive advanced NSCLC.
LBA4 Association of KRAS mutational status with response to pembrolizumab monotherapy given as first-line therapy for PD-L1-positive advanced non-squamous NSCLC in KEYNOTE-042.
Moretti SBVLR, Bellezza G, Metro ASFRG. RET Rearrangement as a Predictor of Unresponsiveness to Immunotherapy in Non-Small Cell Lung Cancer: Report of Two Cases with Review of the Literature.
Role of the dynamic tumor microenvironment in controversies regarding immune checkpoint inhibitors for the treatment of non-small cell lung cancer (NSCLC) with EGFR mutations.
EGFR Mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.
Clin Can Res An Off J Am Assoc Can Res.2016; 22: 4585-4593
Efficacy and safety of programmed cell-Death-Protein-1 and its ligand inhibitors in pretreated patients with epidermal growth-Factor receptor-mutated or anaplastic lymphoma kinase-Translocated lung adenocarcinoma.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial.
Programmed cell death-ligand 1 expression in surgically resected stage I pulmonary adenocarcinoma and its correlation with driver mutations and clinical outcomes.
Clinicopathologic analysis of programmed cell death-1 and programmed cell death-ligand 1 and 2 expressions in pulmonary adenocarcinoma: comparison with histology and driver oncogenic alteration status.
00027-X&id=gr1.jpg){kind=link}
00027-X&id=gr2.jpg){kind=link}