Crizotinib, a small molecule tyrosine kinase inhibitor, was recently approved by the U.S. Food and Drug Administration for the treatment of patients with advanced nonsmall cell lung cancer (NSCLC) that harbor a rearrangement in the anaplastic lymphoma kinase (ALK) gene. First described in NSCLC in 2007, this molecular abnormality is present in approximately 4% to 7% of all cases of lung adenocarcinoma.1, 2 Clinicopathologic characteristics associated with ALK rearrangement include never-smoker status, younger age at diagnosis, and adenocarcinoma with a solid growth pattern and signet ring cells.1 A number of different ALK fusion partners have been reported in NSCLC, all of which ultimately lead to constitutive activation of ALK and its downstream signaling pathways. Aberrant ALK activation results in unrestrained cellular proliferation, evasion of apoptosis, and other phenotypes unique to cancer cells. Although data on the natural history of ALK-positive NSCLC are limited due to the widespread availability of crizotinib, it appears that the disease has a relatively poor outcome in the absence of crizotinib therapy.3
Crizotinib was originally developed to target the c-MET pathway in advanced solid tumors, although it was also known to inhibit ALK. Based on the marked activity of crizotinib in 2 patients with ALK-positive NSCLC, an early-phase study of crizotinib was modified to include an expansion cohort for patients with advanced ALK-positive NSCLC. In this expansion cohort, the objective response rate among 119 patients was 61% and the median progression-free survival was 10 months.4 Similar results were observed in a follow-up phase 2 study of crizotinib involving 136 ALK-positive patients5, 6; in this study, objective response rate was 50% and median response duration was 41.9 weeks. On the basis of its documented response rate, crizotinib received accelerated approval in the United States in August 2011, just 4 years after the initial discovery of ALK rearrangement in NSCLC.
In both phase 1 and 2 testing, crizotinib was tolerated very well. The most common adverse events associated with crizotinib include visual disturbances, gastrointestinal symptoms, peripheral edema, fatigue, decreased appetite, and elevated aminotransferases.4 The majority of the toxicities are grade 1 or 2 in severity, and some improve with continuation of therapy. Based on preliminary patient-reported outcomes from the ongoing phase 2 study of crizotinib, quality of life is maintained during the course of therapy, despite the presence of gastrointestinal and other side effects. In addition, patients have reported a clinically meaningful improvement in disease-related symptoms, including fatigue, insomnia, and appetite loss.
In this issue of Cancer, Weickhardt and colleagues7 report on hypogonadism in male patients treated with crizotinib, a hitherto unreported finding. The investigators measured serum testosterone levels in 19 patients treated with crizotinib. For comparison, they determined testosterone levels in an equal number of patients with NSCLC who were not exposed to crizotinib. The researchers found that testosterone levels were significantly lower in crizotinib-treated patients compared with controls (mean total testosterone: 131 vs 311 ng/dL, respectively). Two patients for whom baseline testosterone levels were measured experienced a significant reduction in testosterone levels within 2 to 3 weeks of initiation of crizotinib therapy. In 4 patients, the authors also tracked testosterone levels in real time as a function of crizotinib exposure; testosterone levels increased rapidly with cessation of crizotinib and decreased rapidly when the drug was restarted. Evaluation of the serum follicle-stimulating hormone and luteinizing hormone levels showed no compensatory increase despite the fall in testosterone, suggesting a potential effect of crizotinib both at a central level in the pituitary and in the testes. The mechanism is not known, but it is interesting to note that c-MET and ALK are both expressed in the central nervous system (CNS) and testes.
The observation of hypogonadism in patients receiving crizotinib therapy is intriguing and raises a number of additional questions. First, the prevalence of hypogonadism appears to be very high in male patients treated with crizotinib. Common symptoms of low testosterone include fatigue, depression, and sometimes insomnia, in addition to sexual dysfunction. Among the 255 patients treated with crizotinib on the phase 1 and 2 studies, treatment-related fatigue was seen in only 20% of patients and was primarily grade 1 or 2. Only 3% of patients had insomnia attributed to study drug, and depression and sexual dysfunction were not among the common adverse reactions (present in ≥10% of patients). It is possible that some of the hypogonadism-related symptoms were underreported or deemed related to the underlying disease. However, it is also possible that patients may not all manifest symptoms of hypogonadism despite a decrease in testosterone level. Second, the authors postulate that crizotinib must exert its effects both in the CNS and in the testes. However, penetration of crizotinib into the CNS is believed to be poor, and in fact, the CNS is the most common site of relapse in crizotinib-resistant patients.8 Understanding the precise mechanism by which crizotinib may suppress testosterone levels is not only of academic interest, but may have implications for the development of second-generation ALK inhibitors, some of which may penetrate into the CNS more efficiently than crizotinib. Finally, the observation of testosterone deficiency in crizotinib-treated males raises the possibility that a similar phenomenon could occur in crizotinib-treated females. Low testosterone levels in females can be associated with similar symptoms as in males, and this question should be addressed.
Hypogonadism occurs quite commonly in patients with advanced cancer, is often underdiagnosed, and is rarely subjected to therapy.9 This study highlights the potential importance of therapy-induced hypogonadism in a subset of patients with lung cancer, and has several immediate clinical implications. First, treating physicians need to be aware of this potential side effect and should look for manifestations of hypogonadism in patients on crizotinib therapy. Importantly, physicians need to inquire specifically about symptoms of testosterone deficiency, including sexual dysfunction, keeping in mind that symptoms of low testosterone can be easily confounded by cancer-related symptoms. As noted by the authors, it is possible that no obvious signal related to testosterone deficiency was detected in crizotinib trials because investigators were unaware of this clinical observation. Second, in those patients who have low testosterone and are symptomatic, it would be reasonable to consider testosterone replacement. Bone density should also be monitored in these patients. Clearly, prospective studies are required to determine whether hormone replacement therapy can alleviate symptoms and improve quality of life in patients with crizotinib-induced testosterone deficiency.
As our knowledge regarding ALK-positive NSCLC grows, it is becoming increasingly clear that crizotinib represents a highly effective and safe treatment option for this group of patients with lung cancer. The improvement in disease-related symptoms and the extended duration of disease control with crizotinib are robust and meaningful to patients. Optimal management of treatable or preventable adverse events can only improve upon the therapeutic index of this agent and help to ensure the best quality of life for our patients.