The Crizotinib Story:
From Target to FDA Approval and Beyond
ALK (anaplastic lymphoma kinase) was first identified as a potential drug target in cancer over 15 years ago when it was discovered as a fusion kinase with nucleophosmin in anaplastic large cell lymphoma.1 However, in the field of lung cancer, ALK was only first recognized as a molecular target four years ago when Dr. Mano and colleagues reported that 6.7% of Japanese patients with NSCLC harbor a fusion of EML4 (echinoderm microtubule associated protein like 4) with the intracellular kinase domain of ALK.2 Based on in vitro studies with an ALK inhibitor WHI-P154 and a Ba/F3 cell line model, Dr. Mano's team suggested that ALK might represent a therapeutic target in patients with ALK-positive NSCLC.2
Fortuitously, a small molecule inhibitor of ALK had just entered the clinic one year prior to Dr. Mano's discovery. PF-02341066, now known as crizotinib, had been developed to target c-MET but was known to also inhibit ALK. In order to test the notion that ALK rearrangement might confer sensitivity to crizotinib and hence represent a bona fide target in NSCLC, a diagnostic assay was needed. This assay was quickly developed by Dr. John Iafrate at Massachusetts General Hospital using commercially available ALK FISH probes. With this assay in hand, investigators proceeded to screen NSCLC tumors, leading to both the identification of the key clinicopathologic features associated with ALK rearrangement,3,4 and screening of clinically enriched populations based on these results. The first patient with advanced, ALK-positive NSCLC was treated with crizotinib at the end of December 2007, just 4 months after the pivotal Nature publication. This patient experienced an almost immediate improvement in his disease-related symptoms, leading to large-scale screening efforts and recruitment of additional ALK-positive patients at phase 1 study sites around the world.5
Efficacy and safety data with crizotinib: basis for accelerated approval in the United States
Results from both the phase 1 and 2 studies have shown that crizotinib is active in patients with advanced, ALK-positive NSCLC. In the phase 1 trial, the objective response rate among 116 evaluable patients was 61%.6 The median time to response was 8 weeks and, as seen with the very first patient treated with crizotinib, rapid and dramatic responses were often noted by investigators. As of June 2011, median progression-free survival (PFS) was 10 months,6 similar to what has been reported with EGFR inhibitors in advanced, EGFR-mutant NSCLC.7 Preliminary results from the phase 2 study were reported at the World Lung Congress in July 2011. Among 133 patients with advanced, ALK-positive NSCLC, the objective response rate was 51% and the disease control rate at 12 weeks was 74%. The follow-up of phase 2 patients was too short to evaluate PFS. The 51% and 61% response rates are impressive since the majority of phase 1 patients and all of the phase 2 patients were previously treated, and studies have shown a response rate of approximately 10% to standard 2nd and 3rd line chemotherapies in unselected patients with advanced NSCLC.8,9
Crizotinib was also shown to have a favorable side effect profile. In both the phase 1 and 2 studies, the major side effects were primarily gastrointestinal (nausea, vomiting, diarrhea, constipation), visual disturbance, and peripheral edema. These were predominantly grade 1. Other notable side effects included fatigue and alteration in taste and/or smell. More significant toxicities included transaminitis and rare cases of drug-induced pneumonitis. Overall, crizotinib was well tolerated in the majority of patients. Based on this safety data, together with the response rates observed in the phase 1 and 2 studies, crizotinib was granted accelerated approval by the FDA in August 2011 for patients with advanced, ALK-positive NSCLC. The approval occurred almost exactly 4 years after the first report of ALK rearrangement in NSCLC.
What about survival benefit with crizotinib?
Overall survival represents the most important indicator of clinical benefit, and is clearly the most reliable endpoint in evaluating a drug's efficacy in a clinical trial. However, the phase 1 trial of crizotinib was not designed to evaluate overall survival. In addition, the phase 3 studies of crizotinib are still accruing patients, and survival results could be confounded by crossover. We therefore pursued a different approach to address whether crizotinib impacts overall survival of ALK-positive NSCLC patients. Specifically, we evaluated survival benefit in a retrospective analysis of survival outcomes between ALK-positive patients on the phase 1 trial, versus ALK-positive patients screened at the same time but treated only with standard therapies for advanced lung cancer. These so-called ALK control patients were also compared with non-ALK patients to address the question of natural history and prognosis associated with ALK rearrangement.10
Consistent with the marked response rates observed in the phase 1 and 2 studies, we found that the overall survival of patients treated with crizotinib on the phase 1 trial was significantly prolonged; 1 year overall survival from starting crizotinib was 74% and 2 year overall survival was 54%. To understand the significance of these numbers, we compared ALK-positive patients treated with crizotinib with ALK-positive patients who had never received crizotinib. For this analysis, we focused on those patients who had received crizotinib as their 2nd or 3rd line therapy, versus those control patients who had received any standard 2nd line therapy. Comparison of survival rates revealed a substantial difference in overall survival, with a hazard ratio of 0.36 in favor of crizotinib, suggesting that crizotinib may improve overall survival in advanced ALK-positive NSCLC10. See Figure 1.
Until recently, the natural history of ALK-positive NSCLC has been poorly understood. This question was relevant to our survival analysis since there was a theoretical possibility that ALK rearrangement by itself might be associated with more indolent disease and a better prognosis, even in the absence of a highly active targeted therapy such as crizotinib. To address this question, we compared the survival of ALK-positive patients who had never received crizotinib with non-ALK patients, specifically patients wildtype for both ALK and EGFR. We found that the overall survival was similar between ALK-positive patients who never received crizotinib and nonEGFR/nonALK patients. Survival was similar even when we took into account demographic features such as age and smoking history, both of which are known to influence prognosis. Thus, in the absence of crizotinib, ALK-positive patients appear to have a similar overall survival as nonALK/nonEGFR patients10. These latter patients represent typical lung cancer patients who generally lack targeted therapy options and have poor survival outcomes.
One of the most important take-home messages from the crizotinib story is the importance of screening patients for ALK rearrangement in order to offer positive patients the opportunity to benefit from a highly effective and safe therapy.
Current recommendations for screening and treating patients with advanced, ALK-positive NSCLC
One of the most important take-home messages from the crizotinib story is the importance of screening patients for ALK rearrangement in order to offer positive patients the opportunity to benefit from a highly effective and safe therapy. At many large academic centers, every patient with advanced lung cancer is screened for ALK rearrangement at the time of diagnosis or initial consultation. Numerous case series have confirmed that ALK positivity is clearly enriched in certain patients, specifically younger patients with a never/light smoking history and adenocarcinoma histology.3,4,11 However, even patients with a significant smoking history have been identified as ALK-positive, and these patients appear to derive the same benefit from crizotinib. Practically speaking, outside the large academic centers where it may not be possible to screen every lung cancer patient for ALK, it would be reasonable to screen all never/light smoking patients, except for those already known to harbor EGFR or KRAS mutations. As mutation testing becomes more widely available, it would certainly be reasonable to think about screening all patients regardless of smoking history, particularly those who are known to be negative for the most common oncogenic drivers.
Like other targeted therapy success stories, patients with ALK-positive NSCLC eventually relapse on crizotinib, typically within one year of initiating crizotinib.
The second equally important take-home message is that patients with advanced, ALK-positive NSCLC should be offered treatment with crizotinib. In the US, crizotinib was recently approved and is now commercially available. Of note, the FDA label does not specify prior therapies. Therefore, in the US, newly diagnosed patients with advanced, ALK-positive NSCLC can be prescribed crizotinib as their first-line therapy. Data supporting the use of crizotinib in the first line setting is limited; in the phase 1 trial, the response rate among 15 patients who received first-line crizotinib was 80%6. However, based on multiple, randomized, first-line trials of EGFR inhibitors in advanced EGFR-mutant NSCLC,12-14 together with the data at hand, it would be reasonable to offer a newly diagnosed ALK-positive patient treatment with crizotinib instead of standard first-line chemotherapy. In the US, newly diagnosed patients can also be referred to the first-line study, PROFILE 1014, which is comparing crizotinib head to head with a platinum/pemetrexed combination in advanced, ALK positive NSCLC. Patients who are initially randomized to standard chemotherapy will be able to crossover to crizotinib at the time of progression. Outside the US, all ALK-positive patients must participate in a clinical trial in order to receive crizotinib. In addition to PROFILE 1014, there is also a 2nd-line registration trial (PROFILE 1007) comparing crizotinib to single agent pemetrexed or docetaxel. ALK-positive patients who are beyond 1st or 2nd line therapy can receive crizotinib through the ongoing phase 2 trial PROFILE 1005. Thus, even outside the US, ALK-positive patients should be able to access crizotinib.
Like other targeted therapy success stories, patients with ALK-positive NSCLC eventually relapse on crizotinib, typically within one year of initiating crizotinib. The development of acquired tyrosine kinase inhibitor (TKI) resistance has emerged as the major hurdle preventing targeted therapies such as crizotinib from having an even more substantial impact on patients. To date, several distinct mechanisms of crizotinib resistance have been identified, including acquisition of a secondary gatekeeper mutation within the ALK TK domain.15 As strategies to overcome resistance clearly depend upon the precise mechanism of resistance, it is imperative that treating oncologists obtain repeat biopsies in relapsing patients for molecular studies. The results of these studies will be crucial to tailoring therapeutic strategies post-crizotinib, and ultimately improving clinical outcomes in patients with advanced, ALK-positive NSCLC.