A “profile” of Xalkori highlights ESMO’s focus on personalized medicine
| Oct 1, 2012
Co-authored by Arnold DuBell, Ph.D., Associate Consultant
During the past few years, drug development for NSCLC has epitomized the term “personalized medicine,” as new patient segments are carved out. A clear example of this is Xalkori® (crizotinib, Pfizer), which targets patients whose tumors harbor EML4-ALK fusion proteins (“ALK+”). In the United States, Xalkori was granted an accelerated approval in August 2011 by the Food and Drug Administration based on results of a Phase I and a Phase II trial. Only recently has Europe acted: Xalkori received a positive opinion from the Committee for Medicinal Products for Human Use in August 2012.
In the second-stage of a two-stage Phase I trial, 116 ALK-positive patients were treated with Xalkori, which produced a response rate of 61%, a median progression-free survival of 10 months, and 81% overall survival at 12 months (Camidge, Abstract 2501, ASCO 2011). Retrospective comparison of this data for the subset of patients treated in the second-line to ALK-positive positive patients who were treated with other second-line chemotherapy suggested an overall survival advantage (Shaw, Lancet Oncol
, 2011). The PROFILE 1005 Phase II study of 400 patients confirmed these data as 83% of patients exhibited target lesion shrinkage (Crinò, Abstract 7514, ASCO 2011). Based on these results, Pfizer initiated the Phase III PROFILE 1007 trial which compared Xalkori to Alimta® (pemetrexed, Eli Lilly) or docetaxel as a second-line therapy in ALK+ advanced NSCLC patients. It was announced in June that this trial achieved its primary endpoint by significantly improving progression-free survival (Pfizer press release, June 19, 2012) and ESMO 2012 offered the presentation of results from this first randomized trial of Xalkori (Shaw, Abstract 1LBA).
The data were impressive in spite of the hype. Compared to chemotherapy, Xalkori more than doubled progression-free survival (Intent-to-treat: 7.7 months versus 3.0 months, HR 0.49, p<0.0001). Subgroup analysis of the control arm indicated that Xalkori performed better than either Alimta (“As treated” population1
: 7.7 months versus 4.2 months, HR 0.59, p=0.0004) or docetaxel (“As treated” population: 7.7 months versus 2.6 months, HR 0.30, p<0.0001). The response rate benefit seen in the earlier Phase I and II trials was confirmed (ORR: 65.3% versus 19.5%, p<0.0001).
The toxicities were expected based on those previously identified. Xalkori administration was associated with an increase in vision disorders, diarrhea, nausea/vomiting, altered liver chemistry, dysgeusia, edema and dizziness. Drug discontinuations were also higher in the Xalkori arm (17% versus 13%). This increased amount of adverse events is tempered by the fact that patients on the Xalkori arm were treated with drug longer: 11 versus 4 median cycles.
An interim analysis of overall survival suggested no survival benefit (20.3 months versus 22.8 months, HR 1.02, p=0.5394); however, the discussant to this presentation (Dr. Soria) noted that 87% of the patients on the control arm crossed over to Xalkori by entering the Phase II PROFILE 1005 trial and that 49% of the patients on the Xalkori arm are still on treatment compared to 16% of patients treated with control. Dr. Soria further put this result in perspective by noting that second-line treatment of Alimta had provided only a 9.3 month overall survival benefit (Hanna, J Clin Oncol
, 2004). Therefore, the use of Xalkori is, as Dr. Soria noted, is “changing the natural history of the disease.” The lack of survival benefit is something being encountered more and more in oncology, and often the post-trial crossover to active drug is felt to be the primary culprit. While researchers and clinicians understand the confounding nature of crossover, and treatment ethics continue to support this practice, the FDA has yet to approve a drug in NSCLC based on PFS benefit without an overall survival benefit. PROFILE 1007 and the ongoing PROFILE 1014 trial in first-line NSCLC are meant to serve as confirmatory trials for Xalkori’s accelerated approval to be converted to full approval. Xalkori may be the first to challenge the FDA’s precedent of overall survival as the approvable endpoint in NSCLC, assuming that further follow-up continues to show no survival benefit in PROFILE 1007. Roche will soon be looking to pose a similar situation to the FDA with Tarceva’s pending regulatory filing for first-line EGFR mutant NSCLC, which also showed an impressive PFS benefit but lacked an overall survival benefit compared to chemotherapy (EURTAC trial; Rosell, Abstract 7503, ASCO 2011).
Today, Xalkori stands all alone, exultant; not only does it act on ALK+ patients, but it may act on other patient segments as well. The lead author of PROFILE 1007, Dr. Shaw, presented a paper at ASCO 2012 suggesting activity of Xalkori in patients whose tumors harbor ROS1 rearrangements (Abstract 7508, ASCO 2012). Given its kinase selectivity profile, one might expect Xalkori to be also tested as a cMet inhibitor (IC50 = 8 nM) and as a RON inhibitor (IC50 = 80 nM). However, competition in the ALK+ segment is lurking. There are several other candidate inhibitors in development. Results of a Phase I trial suggested that LDK378 (Novartis) has marked activity in ALK+ NSCLC, even in patients who have progressed on Xalkori (Shaw, Abstract 440O, ESMO 2012). Additionally, a Chugai / Roche compound, CH5424802, exhibited an 85% overall response rate in relapsed or refractory patients who had not received prior ALK inhibitor therapy (Nishio, Abstract 441O, ESMO 2012).
The “Profile” of Xalkori at ESMO 2012 has cemented the idea of pursuing personalized medicine. Together with the past data on the EGFR inhibitors in patients with EGFR activating mutations, no physician treating a NSCLC patient will reflexively reach for chemotherapy without thinking first of the unique molecular characteristics associated with each individual patient’s tumor. 1. “As treated” population excludes 1 patient in the experimental arm who did not receive crizotinib and 3 patients in the control arm who did not receive chemotherapy treatment