Several interesting lung cancer papers were presented at this year's American Society of Clinical Oncology Annual Meeting. Following are some key presentations that may be of interest to practicing clinicians.
It is now widely recognized that response to epidermal growth factor tyrosine kinase (EGFR TK) inhibitors varies depending on the type of EGFR mutations. There are differences in outcomes even between the two common sensitive EGFR TK mutations. Those with exon 19 deletions do relatively better compared with those with L858R mutation with EGFR TK inhibitors (TKIs) in terms of response rates and progression-free survival.
A number of randomized studies have shown improved response rates and better progression-free survival when patients with EGFR-mutated non-small cell lung cancer (NSCLC) were treated with EGFR TKIs upfront compared with use of cytotoxic chemotherapy.
However, no study has yet shown improved overall survival in patients with EGFR-mutated NSCLC in this setting. Updated results from LUX-Lung 3 and LUX-Lung 6 studies (Abstract 8004) revealed significant differences in outcomes between the two common sensitive EGFR TK mutation groups (exon 19 deletion and L858R) when treated with afatinib.
Both LUX-Lung 3 and LUX-Lung 6 studies randomized patients with previously untreated EGFR-mutant NSCLC to either afatinib or cytotoxic chemotherapy (cisplatin and pemetrexed in LUX-Lung 3 and cisplatin and gemcitabine in LUX-Lung 6). Based on the results of these two studies (primary endpoint: progression-free survival), afatinib has been approved for use in the frontline therapy of EGFR-mutant NSCLC.
The updated data presented at this meeting showed significant improvement (more than one year) in overall survival in patients with exon 19 deletion when treated with afatinib compared with chemotherapy (LUX-Lung 3: 33 vs. 21 months, HR 0.54, p=0.0015; LUX-Lung 6: 31 vs. 18 months, HR 0.64, p=0.0229). There was no improvement in overall survival in the L858R subgroup in either of these two trials when patients were treated with afatinib upfront. Although the combined analysis of patients with common mutations showed a modest (and statistically significant) improvement in overall survival in patients treated with afatinib, most of the benefit seems to be driven by the exon 19-deletion subgroup. Unfortunately there are no data available comparing erlotinib or gefitinib with afatinib in this subgroup.
A Japanese randomized Phase II study (Abstract 8005) presented at this meeting reported significant improvement in progression-free survival in 154 patients with previously untreated EGFR-mutant NSCLC with the combination of bevacizumab and erlotinib compared with use of erlotinib alone (16 vs. 9.7 months, HR 0.54, 95% CI: 0.36-0.79, p=0.0015) with better outcomes in patients with exon 19 deletion compared with L858R.
Not surprisingly, the combination therapy was associated with greater incidence of grade 3 and 4 toxicities (chiefly, hypertension events). The addition of bevacizumab did not impact qualify of life measurably, however. Overall survival data were not available at this time.
These data are somewhat reminiscent of the BeTa trial, where this combination therapy showed some evidence of activity in the EGFR-mutant NSCLC subgroup. An ongoing study in the United States is evaluating this combination in patients with EGFR-mutant NSCLC (NCT01532089).
Three papers presented at the meeting highlighted the promise of the third-generation EGFR TKIs. Despite producing dramatic regressions in EGFR-mutated NSCLC, first-generation EGFR TK inhibitors such as gefitinib and erlotinib eventually become ineffective. AZD 9291 is a third-generation EGFR TKI effective against both “sensitive” (exon 19 deletion, L858R) and T790M involving EGFR TK in pre-clinical models. In a Phase I study with two different cohorts, a dose-escalation one and an expanded cohort, the overall response rate was 51 percent among 177 patients (Abstract 8009). The response rate among 107 patients with centrally confirmed EGFR T790M was an impressive 64 percent, with an overall disease-control rate in this “difficult-to treat” patient population with no standard therapy of 94 percent. Even more strikingly, no dose-limiting toxicities were observed.
Given the fact that this is a mutant-selective inhibitor, normal tissue toxicities (rash and diarrhea) seen typically with the first-generation inhibitors were virtually absent. The response rate with this agent in the T790M-negative group was a modest 23 percent. However, the activity seen in patients in the EGFR T790M group is certainly encouraging.
A similar study used a different compound, CO-1686, a potent oral inhibitor of key activating and T790M resistance mutations and designed to spare wild-type EGFR signaling (Abstract 8010). In a Phase I/II study of 72 patients, an overall response rate of 58 percent was reported in 40 patients with centrally confirmed EGFR T790M. Median progression-free survival was not reached and is estimated to exceed one year.
The third study reported a response rate of 29 percent and a disease-control rate of 75 percent in 48 patients with T790M with HM61713, another EGFR- mutant selective inhibitor (Abstract 8011). Several ongoing clinical trials are now available for patients with EGFR-mutant NSCLC prior to and after therapy with first-generation EGFR TKIs. It is important to biopsy progressing lesions following targeted therapy to identify mechanisms of resistance. It may be possible relatively soon to identify the circulating tumor DNA for these” resistance” mutations.
Updated results from the ASCEND-1 trial of ceritinib (Abstract 8003) in advanced anaplastic lymphoma kinase (ALK+) NSCLC continued to show impressive activity of this agent regardless of whether patients had received an ALK inhibitor previously. Overall response rates were 59 percent for the entire group, 55 percent for those who had received a prior ALK inhibitor (mainly crizotinib), and 66 percent for those who had not received any prior ALK inhibitor.
The progression-free survival rates at one year were 39, 28, and 61 percent, respectively. An important practical point is that 59 percent of patients treated with ceritinib required at least one dose reduction. In a small group of patients (n=14) with measurable brain metastases, the overall response rate was 50 percent. A number of new-generation ALK inhibitors are being studied now in the clinic.
Advanced NSCLC–Biomarker Unselected
Results of two large Phase III studies in molecularly unselected patients with advanced NSCLC may be of interest to practicing clinicians. The SQUIRE study (Abstract 8008) randomized patients with previously untreated squamous type NSCLC to cisplatin and gemcitabine or the same chemotherapy with necitumumab (antibody-targeting EGFR).
The median overall survival improved from 9.9 months with chemotherapy alone to 11.5 months with the addition of necitumumab (HR 0.84, p=0.012); the results were very similar to those of the FLEX study of cisplatin and vinorelbine with or without cetuximab (11.3 vs. 10.1 months, respectively, HR 0.87, p=0.04).
The REVEL study (Abstract LBA8006) randomized patients with advanced NSCLC in the second-line setting to docetaxel with either placebo or ramucirumab. There was an improvement in overall survival with the addition of ramucirumab to docetaxel compared with use of docetaxel alone (10.5 vs. 9.1 months, HR 0.857, p=0.0235). This is perhaps the first time a combination therapy with docetaxel has been shown to be beneficial (albeit, modestly). Unfortunately, there are no clear biomarkers to select patients for either one of these two novel agents.
Surgically Resected NSCLC
The addition of erlotinib postoperatively in the RADIANT study following standard-of-care therapy (surgery with or without adjuvant therapy) did not improve disease-free survival (primary endpoint) or overall survival in patients with stage IB-IIIA NSCLC whose tumors were “positive” for EGFR by immunohistochemistry (Abstract 7501). The median disease-free survival was 48.2 months with placebo and 50.5 months with erlotinib (HR 0.90, p=0.3235).
Of 973 patients, only a small number of patients (n=161) had EGFR mutation. Although the median progression-free survival was superior with erlotinib (46.4 months) compared with placebo (28.5 months), these results were not statistically significant by the design of the study. Unfortunately, this study was not powered or designed to test the hypothesis that patients with EGFR-mutant stage IB-IIII NSCLC would benefit from erlotinib following surgical resection and standard postoperative therapy. A large national study (ALCHEMIST trial) being planned to address the role of molecularly targeted therapy in surgically resected EGFR-mutant NSCLC (with erlotinib or placebo) and ALK-positive NSCLC (crizotinib or placebo) will likely open for enrollment in late summer or early fall of this year.
I am pleased to note the emergence of a number of salvage therapy options for patients with EGFR-mutant and ALK-positive NSCLC. It is highly unlikely that we will make substantial gains with empiric combination therapies without a clearly identified set of biomarkers. I would urge my colleagues to genotype patients with NSCLC on a regular basis at the time of initial diagnosis and when possible at the time of relapse in clearly defined molecular subsets (EGFR mutant, ALK positive) following targeted therapies so that patients can be enrolled in appropriate clinical trials. I do not see a role for targeted therapies in patients with completely resected NSCLC in routine clinical practice.