Exelixis reports cabozantinib phase 2 trial data on bone metastases

Exelixis, Inc. (NASDAQ:EXEL) today reported interim data on pain relief and related reduction in narcotic analgesic use with cabozantinib in castration-resistant prostate cancer (CRPC) patients with bone metastases. Howard Scher, M.D., chief of the Genitourinary Oncology Service at Memorial Sloan-Kettering Cancer Center, will present the data on November 14, 2011 in a poster session at the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics in San Francisco (Abstract B57). The poster will be available at www.exelixis.com/resources/presentations.

Study Results

Effects on pain intensity were measured prospectively in CRPC patients enrolled in the ongoing non-randomized expansion (NRE) cohort of the company's phase 2 randomized discontinuation trial. Sixty-seven patients were enrolled in the NRE cohort at least 6 weeks prior to data cutoff. All patients had bone metastases and 29 (43%) had measurable soft tissue disease. All patients had received prior docetaxel, 16 (24%) had received prior cabazitaxel, 15 (22%) had prior abiraterone or MDV3100, 4 (6%) had prior radionuclides, and 53 (79%) had previously been treated with bisphosphonates and/or denosumab. The median of the average worst pain at baseline determined by the Brief Pain Inventory (BPI) was 3.3 (range 0-7.9). Twenty-nine patients (43%) had moderate to severe pain as evidenced by average worst pain ≥4 at baseline, of which 27 (40%) were taking narcotics for pain. This latter population is the focus of the poster presentation.

Patient compliance with scheduled pain assessment was high; 67 patients completed a total of 268 of 295 (91%) pain assessment intervals. Twenty-nine patients with average worst pain ≥4 at baseline completed 115 of 126 (91%) pain assessment intervals.

Average worst pain improved in 28 of the 29 patients with average worst pain ≥4 at baseline, including 17 patients previously treated with docetaxel alone and 11 previously treated with docetaxel plus abiraterone and/or cabazitaxel. Three of these patients also previously received a systemic radionuclide therapy. Median best pain reduction from baseline was 46%, and 17 (59%) patients had at least a 30% decrease in average worst pain, including patients with 1-3 prior therapies. Of 25 patients with average worst pain ≥4 at baseline and a minimum of 12 weeks of follow-up, 12 (48%) had a durable ≥30% decrease in average worst pain relative to baseline at 2 completed time points which were at least 6 weeks apart. Onset of pain improvement was documented at the first pain assessment at 3 weeks in the majority of patients.

Of the 27 patients with average worst pain ≥4 and taking narcotics at baseline, 15 (56%) decreased their dose by at least 30%, including 7 (26%) who discontinued narcotic drugs completely, 4 (15%) had a stable dose (no change or change ≤30%), and 8 (30%) increased narcotic drug usage.

Pain relief (≥30% pain alleviation relative to baseline) was also accompanied by improved sleep and reduced interference with activities of daily living. Twelve patients with a ≥ 30% pain alleviation at week 12 had a 32% median improvement in sleep and 17% median improvement in interference with activities of daily living. Six patients not experiencing pain relief (<30% improvement in pain relative to baseline) had worsening median interference with sleep (16%) and activities of daily living (20%).

"The improvements in pain observed in this study are significant and suggest that cabozantinib could have an important role in the control of pain related to bone metastases in men with progressive CRPC," said Howard Scher, M.D., an investigator on the trial and chief of the Genitourinary Oncology Service at Memorial Sloan-Kettering Cancer Center. "Cancer-related pain is one of the most feared complications of prostate cancer and a predictor for increased mortality from the disease. New treatment options to relieve and control these debilitating symptoms are urgently needed."

"These prospectively collected data indicate that cabozantinib has a meaningful impact on pain and narcotic use in CRPC patients with bone metastases," said Michael M. Morrissey, Ph.D., president and chief executive officer of Exelixis. "These results importantly corroborate our initial investigator-reported data presented at this year's ASCO meeting, and suggest that cabozantinib can provide meaningful clinical benefit to patients suffering from pain and who have limited treatment options."

Patients in this single-arm study received a starting dose of 100 mg of cabozantinib daily. All patients in the NRE cohort had bone metastases, had received prior docetaxel chemotherapy, and could have received additional systemic therapy including cabazitaxel. Upon study entry, patients were required to have demonstrated objective evidence of disease progression during or within 6 months of docetaxel or cabazitaxel therapy.

Pain and Analgesic Medication Assessment

Pain was assessed using an Interactive Voice Response System comprising a 4-item questionnaire. The questionnaire assessed pain at its worst, disturbed sleep at its worst, and interference with activities of daily living, all within the previous 24 hours. These three items were measured using an 11-point rating scale (0-10), in which 0 indicates absence of the symptom. The worst pain intensity rating is derived from the BPI, which is widely used and accepted in contemporary pain studies (Atkinson et al., 2010). Changes in a patient's pain were assessed relative to their baseline pain score. Narcotic analgesic medication use was assessed by Patient Diary. All pain and analgesic medication use assessments were reported as the average daily score over a 7-day interval. The baseline assessment was taken within 14 days prior to the first dose of cabozantinib, while the post-baseline assessments were taken at Week 3, Week 6, and every 6 weeks thereafter. For the assessment interval to be considered valid, the patient had to complete a minimum of 4 of the 7 questionnaires administered during the 7-day assessment interval.

Changes in pain assessment were evaluated in patients with moderate-to-severe baseline pain, defined as an average daily worst pain score ≥4. Patients were evaluated for achieving a ≥30% decrease in average daily worst pain relative to baseline. Patients were also evaluated for changes in narcotic analgesic use relative to baseline.

Safety and anti-tumor activity in the NRE cohort of CRPC patients will be summarized in a future presentation.