Small cell lung cancer (SCLC) is estimated to cause 250,000 deaths per annum [1], is strongly linked to tobacco smoking and is predicted to increase in incidence globally by 1.59% annually through 2024 [2]. At the 2017 IASLC World Conference on Lung cancer held in Yokohama, Japan, researchers presented papers that provide new insights towards improving patient outcomes.
Current treatment patterns for patients with SCLC
The standard of care (SOC) for first line treatment of extensive disease SCLC has been platinum and etoposide for several decades, but while the majority of patients initially respond, relapse is invariable, with limited second line treatment options. A real world outcomes study of almost 6000 patients in 2014-2016 found that only 19% of patients receive a second line of treatment. Furthermore, around half of patients, regardless of geographic location, progress within 3-6 months of initial treatment. [3] These findings reinforce the high unmet need for better therapies in both first and second line settings.
Biomarkers for treatment selection
Immunotherapy is emerging to be a viable treatment for SCLC. Initial findings from CheckMate 032, a phase 1/2 clinical trial that evaluated nivolumab with or without ipilimumab in solid tumors, demonstrated efficacy in patients with previously treated SCLC [4]. The response rates were in the region of 10-30%; therefore, the identification and use of objective molecular tests to guide patient selection is needed to optimise chance of benefit. Programmed death ligand 1 (PD-L1), was not found to be predictive for response in the checkmate 032 study. In contrast, the results from an exploratory analysis designed to assess tumor mutation burden (TMB) are promising. Whole exome sequencing (WES) was performed on tumors and matched blood samples to group patients equally into low, medium and high TMB cohorts. Among patients with an evaluable TMB result who were treated with only nivolumab, objective response rate (ORR), progression-free survival (PFS) and overall survival (OS) were improved in the high TMB cohort as compared to the medium and low TMB cohorts. Furthermore, in those treated with nivolumab and ipilimumab, the ORR and the 1-year survival rates were 46.2% and 62.4%, respectively, in patients with a high TMB. [5] The results provide a clear signal that TMB warrants further evaluation as a predictive biomarker in ongoing trials of immunotherapy in patients with SCLC.
Rovalpituzumab tesirine (Rova-T) is a promising DLL3-targeted antibody-drug conjugate that has entered phase III clinical trials in various treatment settings for SCLC. Tissue expression of DLL3 by immunohistochemistry is currently used as a biomarker for patient selection. In SCLC, where small biopsies are characteristic, the question of how representative a small diagnostic biopsy is of DLL3 expression in the whole tumor is therefore relevant. In a study of 43 cases where a resected tumor and biopsy were both available, Dr Yan and colleagues demonstrate the biopsy to be reliable for DLL-3 expression. Also of interest, in a series of 335 SCLC patients, high DLL-3 expression was associated with smoking history, expression of TTF-1 (a routinely used diagnostic biomarker for lung cancer) and poor survival trends. The results raise the possibility that negative TTF-1 expression could be used as a surrogate for negative DLL-3 expression [6] and that other biomarkers alongside DLL3 should be prospectively examined in ongoing studies.
A further study examined the impact of baseline immune suppression and systemic inflammation assessed using routinely monitored hematologic markers; total lymphocyte count (TLC) and neutrophil-to-lymphocyte ratio (NLR) on OS in patients with extensive stage -SCLC. In 253 patients undergoing SOC treatment, the median OS was noted to be significantly worse in patients with lower pretreatment TLC and higher pretreatment NLR. These findings are intriguing, with important implications for stratifying patients in clinical trials, particularly of immunotherapy approaches. [7]
Liquid biopsies as a surrogate for tissue biopsy
As previously mentioned, a major challenge for precision medicine approaches that require detailed molecular analysis and/or immunohistochemistry panels is that SCLC is usually diagnosed from a small biopsy or cytology specimen. As a case in point, only 51% of patients had tissue of sufficient quantity and quality for analysis of biomarkers in a randomised phase II trial of a PARP-1 inhibitor presented in Yokohama[8]. Alternatively, a blood sample is readily obtained and circulating tumor DNA (ctDNA) is an attractive, but as yet unvalidated, prospect for patients with SCLC. In Yokohama, two studies were presented that provide the first evidence for the feasibility of liquid biopsy as a clinical tool in patients with SCLC. Dr Lam and colleagues applied targeted-capture deep sequencing of 545 cancer genes to ctDNA samples collected before any treatment then serially during treatment from 23 patients with SCLC[9]. In 8 patients it was possible to sequence and compare the pretreatment tumor biopsies for the same gene panel. Mutations were identified in all 44 ctDNA samples with a median of 16 mutations per sample. The most frequently mutated genes were TP53 (in 91% of cases) and RB1 (in 65% of cases) consistent with what is known about the somatic genetics of SCLC. There was good agreement (93.2%) between the mutations identified in the tumor biopsies and ctDNA from the same patients. The authors noted substantial genomic heterogeneity and found that the quantity (measured by variant allele frequency (VAF)) of clonal mutations before any treatment, was associated with survival independent of stage, age, or platinum sensitivity. The median OS was 8.1 months (95% CI, 5.5 to 10.7 months) versus 24.9 months (95% CI, 0.0 to 51.2 months) in patients with higher versus lower than median clonal-VAF, respectively (p=0.004). Analysis of serial ctDNA before and during treatment also showed that clonal-VAF closely tracked with treatment responses. Similarly, Dr Lange et al demonstrated that longitudinal sampling of ctDNA can mirror response in patients with SCLC and that a rise in VAF (quantity) of one mutation was detected before radiological progression. These ‘liquid biopsy’ studies are an important advance in technology application for patients with SCLC, and open the doors to further evaluation of ctDNA in the context of standard and novel drug treatment and push us towards the ability to select patients for treatment on the basis of molecular characteristics of the tumor.
Conclusion
SCLC has a 2-year survival of less than 10% and has been designated a research priority under the 2013/14 US recalcitrant cancer act. While there is much work to do, collaboration between clinicians and scientists to address the unmet need in SCLC is accelerating and on the right track towards new approaches that harness genomic technologies and new understanding of SCLC biology.