Beyond chemotherapy - emerging biomarkers and therapies as SCLC enters the immune checkpoint era

Abstract

For decades, clinicians have seen no therapeutic advances for SCLC, including no FDA-approved targeted therapies; recently, immune checkpoint blockade has emerged as a promising new option for the treatment of relapsed SCLC (including recent FDA approval of nivolumab in the third-line setting) and may soon represent the frontline standard of care in combination with chemotherapy. However, we need to uncover biomarkers to guide patient selection and develop novel approaches to enhance response to immunotherapies.

Over the past 15 years, targeted and immune-based therapies have transformed the treatment of non-small cell lung cancer (NSCLC), driving a corresponding revolution in biomarker testing, as current guidelines recommend testing all stage IV patients for actionable gene alterations (EGFR, ALK, BRAF and ROS1) and PDL1 expression to optimize treatment selection (1). This year, pivotal Phase III trials (Keynote-189 and −407, IMpower-131 and −150) have reshaped frontline treatment for patients, as immunotherapy – either alone or with chemotherapy – emerges as a new frontline standard for all patients with advanced driver-negative NSCLC (2–6).

In stark contrast, therapeutic advances have been elusive for small cell lung cancer (SCLC). SCLC is the most aggressive form of lung cancer, accounting for 15% of lung cancers in the United States (7). For several decades, systemic treatment for extensive- and limited-stage SCLC has relied upon the one-regimen-fits-all approach of platinum plus etoposide chemotherapy (8). Although most patients have robust responses initially, the majority relapse within months, contributing to a dismal 5-year overall survival (OS) of ≤7% (7).

Unlike NSCLC, there were no FDA-approved targeted therapies for SCLC until August 2018, when the anti-PD1 monoclonal antibody nivolumab was approved by the FDA for patients who have received two or more prior lines of therapy. The recommendation of immunotherapy with nivolumab (anti-PD1) alone or in combination with ipilimumab (anti-CTLA4) for relapsed disease had been introduced in 2016 to the NCCN’s SCLC guidelines (8) on the basis of results from the phase I/II CheckMate032 trial (9, 10). In relapsed SCLC patients, objective response rates (ORR) were 11% (nivolumab) and 22% (nivolumab + ipilimumab), while 2-year OS was 14% with monotherapy and 26% with combination therapy. Notably, among those receiving clinical benefit were patients with platinum-resistant and/or heavily pretreated SCLC, a population typified by therapeutic resistance. While response rates were improved with combination immune checkpoint blockade, it should be noted that adverse events (AEs) were higher with nivolumab + ipilimumab, with 33% of grade 3/4 AEs versus 14% with nivolumab, including myasthenia gravis, renal failure and immune-related pneumonitis and encephalitis (10). More recently, preliminary data from a phase I study evaluating the combination of durvalumab (anti-PDL1) and tremelimumab (anti-CTLA4) in relapsed SCLC confirmed a tolerable profile and 1-year OS of 41.7% (11). These data suggest encouraging activity in highly refractory populations and signal a promising future for immunotherapy in SCLC.

Surprisingly, CheckMate032 did not support the use of PDL1 as a biomarker in SCLC. PDL1 expression was rare (observed in only 17%) (10, 12) and – in contrast to NSCLC – clinical benefit was independent of PDL1 expression by automated assessed tumor positive score as ORR in PDL1-negative patients was 14% and 32.3% versus 9.1% and 10% in PDL1-positive patients with nivolumab (n=245) and nivolumab plus ipilimumab (n=156), respectively (10, 12). Contrastingly, preliminary data from Keynote-158, an ongoing phase II single-arm trial of pembrolizumab (anti-PD1) in relapsed SCLC patients, suggested a higher ORR and OS in PDL1-positive patients, but no clear difference in progression free survival (PFS) by PDL1-status (13). However, the latter trial employed a combined score to assess PDL1 positivity that included both tumor and stromal compartments, which may underlie some of the apparent discrepancy. Although these data will likely continue to evolve, these existing results currently do not support use of PDL1 IHC as a method for SCLC-patient selection.

In contrast to PDL1 expression, Hellmann and colleagues have recently demonstrated that tumor mutational burden (TMB) may be an alternative predictive biomarker for clinical benefit from immunotherapy in SCLC patients (12). This is consistent with recent observations from NSCLC, where TMB was also found to predict ORR and PFS with nivolumab plus ipilimumab treatment --independent of PDL1 expression-- in multivariate analyses (6). In a retrospective analysis, TMB was calculated for SCLC patients from Checkmate032 with sufficient tissue for whole-exome sequencing (WES) as the total number of somatic missense mutations, with patients divided into 3 groups [TMB-high (>248), TMB-medium (143–247), or TMB-low (0–142)] (12). In patients with TMB-high tumors, 1-year OS was 35.2% with nivolumab and almost doubled (62.4%) with nivolumab plus ipilimumab, but only ~20% in both treatment arms with TMB-medium/low. ORR were also 2–3 times higher in TMB-high patients treated with combination therapy (46.2% in TMB-high versus 16% and 22.2% in TMB-medium/low) or with nivolumab alone (21.3% in TMB-high versus 6.8% and 4.8% in TMB-medium/low) (13). The above results clearly demonstrate that for unselected relapsed SCLC populations, treatment with anti-PD1 +/− anti-CTLA4 therapies compares favorably to historical outcomes for chemotherapy. Nevertheless, even among TMB-high patients, less than half of patients respond to immunotherapy, emphasizing the need for novel applications and combinations of available therapies and for predictive biomarkers to optimize SCLC-patient selection for immunotherapy.

At the time of our writing, a press release has reported that the randomized, phase III IMpower133 trial comparing carboplatin-etoposide plus placebo or plus atezolizumab (anti-PDL1) as first-line therapy, followed by atezolizumab maintenance, in unselected SCLC patients met its pre-specified co-primary endpoints of improved PFS and OS. It is very likely that – similar to NSCLC – this chemo-immunotherapy combination will become the new frontline standard of care. However, several key issues remain undecided, including the role of TMB or PDL1 expression in selecting frontline therapy and the utility of chemo- or targeted-therapy combinations with immunotherapy to overcome primary or acquired immunotherapy resistance.

Preclinical work from our lab and several others suggest that pharmacologic inhibition of DNA damage response (DDR) can lead to changes that may promote response to immune checkpoint blockade, including increasing neo-antigens, enhancing PDL1 expression, and attracting/activating T-lymphocytes (14, 15). These data suggest the need for developing clinical strategies involving concurrent or sequential targeting of DDR and immune checkpoints. Elsewhere, in addition to immune checkpoint blockade, promising new therapeutic strategies for SCLC are under investigation (16–18), including DLL3-targeting approaches, with two new immunotherapeutic approaches, DLL3-targeting chimeric antigen receptor (CAR) T-cells and a DLL3 bispecific T-cell engager (DLL3-BiTE), both in early phases of clinical evaluation (19, 20).

In conclusion, immune checkpoint blockade has emerged as a promising treatment option for relapsed/refractory SCLC patients, as we await results from randomized, controlled trials to determine if immunotherapy +/− chemotherapy is indeed superior to current standards. It remains to be seen whether biomarkers such as PDL1 expression and TMB will define the populations with greatest benefit from these therapies, as they have in NSCLC. As clinicians prepare for the new standard of care in SCLC, preclinical and early-phase clinical researchers seek to discover novel approaches to enhance response to existing immunotherapies, along with developing alternative immunotherapeutic strategies, including CAR T-cells. As more intriguing data emerge, so does the possibility that clinicians treating SCLC may finally experience the luxury, and challenges, of choice.