Recent advances in squamous non-small cell lung cancer: evidence beyond predictive biomarkers

During the last decade, therapeutic options for advanced non-small cell lung cancer (NSCLC) have grown significantly. However, these achievements have been mostly limited to the non-squamous cell carcinoma (non-SCC), leaving squamous cell carcinoma (SCC) with high demand for new treatments. Not only does SCC preclude access to pemetrexed or bevacizumab for poor efficacy or safety, but it also lacks predictive molecular biomarkers such as activating mutations of the epidermal growth factor receptor (EGFR). The mutation burden is also higher in SCC, although no suitable target-mutation has been identified so far. Additionally, SCC is highly immunogenic, thus susceptible to immunotherapeutic approaches. Both characteristics are related to cigarette smoking, which is commonly associated with SCC and with lack of driver mutations in non-SCC.[Citation1,Citation2] Interestingly, the last few months have seen a partial inversion of the general trend favoring non-SCC.

In first-line, an unexpected twist favoring the SCC has been recently observed with necitumumab, a monoclonal antibody targeting EGFR. Its clinical efficacy in combination with first-line chemotherapy was evaluated in two Phase III trials, the first (INSPIRE) involving non-SCC and cisplatin-pemetrexed as the regimen of choice, the second (SQUIRE) involving SCC and employing cisplatin-gemcitabine. The rationale for treating both histologies relies in the fact that, while EGFR gene mutations are more common in non-SCC, the actual receptor can be over-expressed in both histologic subgroups (being actually more often over-expressed in SCC) [Citation3]; additionally, data from the FLEX trial involving cetuximab and first-line chemotherapy for NSCLC had already implied that EGFR-antibodies might be a suitable approach for both histologies.[Citation4] Surprisingly, adding necitumumab did not improve overall survival (OS; 11.3 vs. 11.5 months, p = 0.96) or progression-free survival (PFS; 5.6 vs. 5.6 months, p = 0.66) in non-SCC; additionally, the experimental arm was flawed by an unacceptable increase of thrombosis-related severe adverse events (INSPIRE).[Citation5] On the contrary, necitumumab improved significantly, albeit modestly, survival in SCC (OS: 11.5 vs. 9.9 months, p = 0.01; PFS: 5.7 vs. 5.5 months, p = 0.02); additionally, its safety profile, while characterized by increased thromboembolic events, was considered generally manageable (SQUIRE). Notably, not only SCC suffered less toxicity than non-SCC, but it achieved a significant survival benefit; while such differences might be actually related to the different regimens of choice, rather than to histology, the results are nonetheless meaningful and, should necitumumab receive the approval from the Food and Drug Administration (FDA), it would be the first targeted agent specifically registered for first-line in SCC. Although further sub-group analyses are still awaited, no suitable predictive biomarker for necitumumab has been identified so far and the categorical division between high and low EGFR protein expression (measured as H-score) failed to predict the outcomes in both trials, in contrast with the retrospective evidence with cetuximab [Citation6]; however, this result may depend from the specific criteria employed.

Even more impressive results involved second-line, where the current standard has been represented for long time by docetaxel monotherapy, which was registered on the basis of improved OS over placebo (7.0 vs. 4.6 months; p = 0.047).[Citation7] The programmed-death protein 1 (PD-1) inhibitor nivolumab (an immune check-point inhibitor) has been successfully compared to docetaxel as second-line for advanced NSCLC in the Phase III trials CheckMate 017 (SCC; 352 patients) and CheckMate 057 (non-SCC; 582 patients). In CheckMate 017, nivolumab showed a significant advantage in terms of OS (9.2 vs. 6.0 months, p < 0.001) and PFS (3.5 vs. 2.8 months, p < 0.001) [Citation8]; in the interim analysis of CheckMate 057, nivolumab achieved improved OS (12.2 vs. 9.4 months, p = 0.002), while PFS was not prolonged (2.3 vs. 4.2 months, p = 0.39).[Citation9] Following such data, nivolumab has been rapidly approved in second-line for SCC [Citation10]; additionally, most recently, the approval has been extended also to non-SCC, based on the OS advantage.[Citation11] The expression of PD-1 ligand (PD-L1) was evaluated in both trials and, interestingly, it was neither prognostic nor predictive in CheckMate 017, while the interim analysis of CheckMate 057 showed a strong predictive association between PD-L1 expression and clinical outcome. These findings are partially similar to the results of pembrolizumab, another PD-1 inhibitor, in a Phase I trial (KEYNOTE-001) including 495 previously treated and treatment-naïve patients with advanced NSCLC (SCC and non-SCC). The outcomes were impressive, with OS reaching 16.2 months in previously untreated patients, and consistent between histologies; moreover, the increased expression of PD-L1 was strongly correlated with objective response, and this finding, unlike CheckMate 017, was observed also in SCC.[Citation12] Recently, FDA approved pembrolizumab in previously treated NSCLC irrespective of histology, paired with a companion PD-L1 expression test.[Citation13] This inconsistency with nivolumab might be at least partially correlated with the use of different PD-L1-testing assays across the different studies; a standardization of such assessment, while challenging, needs to be pursued in order to acquire suitable data for use in clinical practice. Of note, it has been suggested that the mutational burden within the tumor, consisting in genetic and epigenetic alterations, might result in the development of altered antigenic profiles recognizable by immune cells as non-self; hence, such signatures, as well as mutational burden itself, might represent suitable biomarkers for immunotherapy in NSCLC.[Citation14]

While immunotherapy has the qualities for outclassing docetaxel as the new standard for second-line, chemotherapy is still a valuable option for selected, fit patients, eventually keeping immunotherapy for further lines; the most notable attempt in order to improve the efficacy of chemotherapy is observed in the REVEL trial. In this study, 1253 NSCLC patients were randomized to receive docetaxel plus placebo or plus ramucirumab, an anti-vascular endothelial growth factor receptor-2 (VEGFR-2) monoclonal antibody, in second-line; the experimental arm achieved longer OS (10.5 vs. 9.8 months, p = 0.023) and PFS (4.5 vs. 3.0 months, p < 0.0001).[Citation15] Notably, the trial design excluded patients with altered coagulation function or proteinuria, consistently with the drug class profile, while SCC was included provided that major airway or blood vessel invasions, or tumor cavitation, had been excluded at baseline imaging; as a result, 328 patients with SCC (26% of the total) were enrolled. The experimental arm was characterized by an increased incidence of hemorrhagic events, epistaxis, and hypertension, which represent the trademark toxicity for VEGF-targeting agents; however, unlike bevacizumab toxicities, they were not more frequent in SCC. Additionally, the benefits in terms of OS, PFS, and objective response rate (ORR) in the experimental arm were consistent between histologies. Although the study was not actually powered for subgroup analysis, REVEL suggests that, while angiogenesis had been considered for long time an approach restricted to non-SCC, VEGF-disrupting agents can be employed with benefit and acceptable safety in both histo-types, provided that a radiologic screening is performed. Recently, FDA approved ramucirumab in combination with docetaxel as second-line in NSCLC [Citation16] and, notably, no molecular biomarkers are known for this drug.

Another attempt to improve the efficacy of docetaxel involved the addition of nintedanib, a multi-target inhibitor targeting VEGFR, fibroblast growth factor receptor (FGFR) family, and platelet-derived growth factor receptor (PDGFR), in the LUME-Lung 1 trial; although the addition of nintedanib improved PFS in both histologies (3.4 vs. 2.7 months; p = 0.0019), only patients affected by non-SCC achieved an improved OS, while SCC (555 patients; 42%) showed no OS difference (p = 0.8907).[Citation17]

Another relevant result in previously treated SCC comes from the Phase III LUX-Lung 8 trial, designed to compare an irreversible EGFR inhibitor, afatinib, and the reversible inhibitor erlotinib as second-line for SCC patients unselected for EGFR mutations (398 patients receiving afatinib and 397 receiving erlotinib, respectively). Afatinib was superior to erlotinib in terms of PFS (2.4 vs. 1.9 months, p = 0.0427), OS (primary analysis: 7.9 vs. 6.8 months, p = 0.0103), and disease control rate (51% vs. 40%, p = 0.0020); as expected, drug-specific adverse events (gastrointestinal and skin toxicities) were more frequent with afatinib, but its safety profile was generally manageable.[Citation18] The results achieved by afatinib effectively revamp the role of EGFR inhibitors beyond the presence of activating mutations in pre-treated SCC.

In conclusion, the availability of new agents able to increase survival has brought a much-needed update for SCC, enabling it to keep the pace with non-SCC; nonetheless, the identification of reliable and consistent biomarkers still represents a top priority in SCC research, as they might be able to predict which patients are really going to benefit the most and can reasonably face the risk of increased adverse events. With reference on biomarker-based therapies for SCC, a promising approach in SCC involves targeting FGFR family, and particularly FGFR1, which is more frequently amplified in SCC (22%) than non-SCC (3%); dovitinib, a multi-target inhibitor active on FGFR family among others, as well as the selective FGFR inhibitors AZD4547 and BGJ398 are currently being studied in Phase II/III clinical trials after having achieved noteworthy results in pre-clinical studies and objective responses in FGFR1-amplified SCC patients in Phase I trials.[Citation19] Should such trials be successful, another therapeutic approach might soon become available for SCC.

Financial & competing interests disclosure

Carlo Genova participated in speaker’s bureau for Boehringer Ingelheim and received honoraria from Bristol Myers Squibb and ItalFarmaco. Erika Rijavec participated in speaker’s bureau and advisory boards for Boehringer Ingelheim and received honoraria from Bristol Myers Squibb. Francesco Grossi participated in speaker’s bureau for Boehringer Ingelheim and participated in advisory boards and lectures for the following companies: Amgen, Astra Zeneca, Bristol Myers Squibb, Boehringer Ingelheim, Celgene, Eli Lilly, Merck Sharp & Dohme, Novartis, Pierre Fabre, Pfizer, Roche. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.