Castration-resistant prostate cancer: Cabozantinib linked to high rates of complete or partial bone scan resolution, pain relief, improvements in hemoglobin, and tumor regression
Exelixis has reported updated interim data from the cohort of patients with metastatic castration-resistant prostate cancer ( CRPC ) treated with Cabozantinib ( XL184; Cometriq ) in an ongoing phase 2 adaptive randomized discontinuation trial. The data support the findings that Cabozantinib reduces or stabilizes metastatic bone lesions in nearly all patients evaluable by bone scan, reduces bone pain and narcotic analgesic medication and results in an increase in hemoglobin in patients with anemia. The data also suggest that Cabozantinib shows an encouraging early signal of durable clinical benefit in both Docetaxel-naïve and pretreated patients.
Cabozantinib, an inhibitor of tumor growth, metastasis and angiogenesis, simultaneously targets MET and VEGFR2, key kinases involved in the development and progression of many cancers. Prominent expression of MET has been observed in primary and metastatic prostate carcinomas, with evidence for higher levels of expression in bone metastases. Overexpression of hepatocyte growth factor ( HGF ), the ligand for MET, has also been observed in prostate carcinoma, and increased plasma levels of HGF are associated with decreased overall survival in CRPC. Data from preclinical studies also suggest that both HGF and MET are regulated by the androgen signaling pathway in prostate cancer, where upregulation of MET signaling is associated with the transition to androgen-independent tumor growth. Additionally, both the MET and VEGFR signaling pathways also appear to play important roles in the function of osteoblasts and osteoclasts, cells in the bone microenvironment that are often dysregulated during the establishment and progression of bone metastases.
The primary cause of morbidity and mortality in patients with castration-resistant prostate cancer is metastasis to the bone, which occurs in about 90% of cases. Bone metastases cause local disruption of normal bone remodeling, with lesions generally showing a propensity for an osteoblastic ( bone-forming ) phenotype on imaging. These lesions often lead to increased skeletal fractures, spinal cord compression, and severe bone pain. Osteoblastic lesions are typically visualized in CRPC patients by bone scan, which detects rapid incorporation of 99mTc-labeled methylene-diphosphonate radiotracer into newly forming bone. In addition, increased blood levels of ALP and CTx, markers for osteoblast and osteoclast activity, respectively, are often observed in castration-resistant prostate cancer patients with bone metastases, and are associated with shorter overall survival.
As of the January 27, 2011 cut-off date, accrual in this cohort was complete at 168 patients, randomization was halted, and randomized patients were unblinded based on an observed high rate of clinical activity.
The efficacy-evaluable population for tumor response per mRECIST includes 100 patients with at least 12 weeks of follow-up. The bone scan evaluable population includes 62 patients with a baseline bone scan, evidence of bone metastasis, and at least 1 post-baseline assessment. All patients had measurable soft tissue disease, approximately half ( 47% ) had evidence of visceral disease ( e.g., liver or lung ), and approximately half ( 47% ) were Docetaxel-pretreated.
Complete or partial bone scan resolution - Of 62 patients evaluable by bone scan, 53 ( 85% ) achieved either complete or partial resolution of metastatic lesions on bone scan by independent radiologist review. Eight other patients ( 13% ) had stable disease on bone scan, resulting in an overall rate of disease control in bone of 98% ( 61/62 ). Only one patient ( 2% ) had progressive disease as their best assessment. High rates of bone scan resolution were observed in both Docetaxel-pretreated and Docetaxel-naïve subgroups.
Hemoglobin in patients with anemia - Castration-resistant prostate cancer patients with bone metastases frequently experience anemia, which in prior studies has been associated with a worse prognosis. In this study, the majority of patients with anemia ( hemoglobin less than 11g/dL ) had sustained treatment-emergent increases in hemoglobin as compared with baseline. The median maximum rise in hemoglobin in anemic patients ( Hb less than 11 g/dL ) was 2.2 g/dL ( range, 0.6-3.5 ).
Progression-free survival - The median follow-up for all 100 patients is 3.8 months ( range, 0.7 to 15.2 months ). In the Randomized Discontinuation Phase, a total of 31 patients with stable disease at week 12 were randomized to either placebo or Cabozantinib. The median progression free survival ( PFS ), assessed by investigators, is 40 days ( 95% confidence interval: 37, 82 days ) for the placebo group ( n=17 ), while median PFS data for the Cabozantinib group ( n=14 ) are not yet mature with a censoring rate of 79%.
For the overall population, excluding those randomized to placebo, the median progression-free survival data for patients continuously treated with Cabozantinib are not yet mature with censoring rates of 85% and 79% for the Docetaxel-naïve and -pretreated populations, respectively. However, progression-free survival appears to be independent of prior Docetaxel therapy, which historically has conveyed a worse prognosis.
Symptomatic bone pain - Data on bone pain and narcotic use, as assessed by the investigator, were collected retrospectively. A total of 43 patients had bone metastases and bone pain reported at baseline, and at least one post-baseline assessment of pain status. Of these patients, 26 ( 60% ) had pain improvement at either week 6 or 12. Narcotic analgesic medication was required at baseline for control of bone pain in 28 patients assessable for post-baseline review of narcotic consumption. Thirteen of 28 ( 46% ) patients were able to decrease or discontinue narcotic medication for bone pain.
Soft tissue lesions - One hundred patients were evaluable for response by mRECIST. The overall week-12 disease control rate ( stable disease plus partial response [ PR] ) was 74%. Tumor shrinkage was observed in 61 of 91 patients ( 67% ) with measurable soft-tissue metastatic lesions and at least one post-baseline scan. To date, 6 of 100 patients ( 6% ) evaluable by mRECIST achieved a confirmed PR. Stable disease was reported in 82 patients ( 82% ), and 2 other patients have had unconfirmed PRs awaiting confirmation. Prostate-specific antigen ( PSA ) changes were observed to be independent of reduction or stability of tumor target lesions, resolution of bone lesions on bone scan, and changes in bone pain.
Markers of bone formation and resorption - Alkaline phosphatase ( ALP ) and type I collagen C-telopeptides ( CTx ), which are markers of osteoblast ( bone formation ) and osteoclast ( bone resorption ) activity, respectively, are often elevated in patients with bone metastases. Reductions in levels of CTx and ALP have been correlated in the past with a reduced risk of skeletal-related events and mortality, respectively. Of 16 patients treated with Cabozantinib who had ALP levels at least twice the upper limit of normal, known bone metastases, and at least 12 weeks of follow-up, 15 had decreases in ALP. Similarly, 42 of 48 patients with known metastases experienced a decrease in CTx, regardless of prior bisphosphonate status. ALP and CTx levels decreased from baseline by approximately 60% at weeks 12 and 24, respectively.
Safety and tolerability - Safety data are available for the lead in phase of the study for 100 patients with at least 12 weeks of follow-up. The most common greater than or equal to grade 3 adverse events, regardless of causality were fatigue ( 15% ), hypertension ( 8% ), PPE syndrome ( 5% ), back pain ( 3% ), decreased appetite, nausea, vomiting, rash, hemorrhage, abdominal pain ( each 2% ), diarrhea, dyspnea, and cough ( each 1% ). No Cabozantinib-related grade 5 adverse events were reported. At least one dose reduction was reported in 51% of patients. ( Xagena )
Source: Exelixis, 2011