CHICAGO, IL USA (UroToday.com) - Dr. Dawn Hershman of Columbia University discussed outcomes from the SWOG S9346 study evaluating men with metastatic prostate cancer receiving intermittent vs continuous androgen deprivation therapy (ADT).
She reviewed the data from multiple population-based cohort studies that demonstrate the association between ADT and sexual dysfunction, osteoporosis/fractures, cardiovascular disease, diabetes, and cognitive changes. S9346 was a prospective randomized clinical trial which allowed prospective data collection on patients randomly assigned to treatment. This type of data, though limited by including a more select population than a SEER-Medicare analysis, avoids selection bias and allows assessment of more detailed treatment and prognostic data. S9346 merged the two by linking with Medicare claims data. The investigators sought to evaluate the late effects of treatment with ADT among men randomized to intermittent vs continuous ADT for metastatic prostate cancer. They hypothesized that long term cardiovascular and endocrine effects would be less among men treated with intermittent ADT.
S9346 was a non-inferiority design, randomized, controlled trial that included men with metastatic prostate cancer. Men were randomized to treatment with continuous ADT vs intermittent ADT (if PSA ≤ 4 during months 6 and 7, men stopped ADT and restarted when PSA increased to baseline PSA or if men developed symptoms). Intermittent ADT in the original report failed to reach the non-inferiority threshold, with median overall survival in the continuous arm being 5.8 months vs 5.1 months on the intermittent arm. The study also evaluated health-related quality of life outcomes, including mental health, physical function, erectile function, and libido at 0, 3, 9, and 15 months, and there were minimal differences in these over the course of the study. Adverse events were similar.
This study assessed long-term cardiovascular and endocrine outcomes, including ischemic and thrombotic events (myocardial infarction, ischemic heart disease, and DVT/pulmonary embolus), diabetes hypercholesterolemia, obesity, osteoporosis, fracture, sexual dysfunction, dementia, and depression. The investigators performed a time to event analysis and used a multivariable Cox regression to assess the impact of treatment on those outcomes. The study groups were well balanced in terms of age, race, BMI, comorbidity at baseline, and prostate cancer severity. During the course of the study, there was a higher rate of ischemic and thrombotic events in the intermittent arm when compared to the continuous treatment arm (HR 0.68, p=0.02 for all events, and HR=0.55, p=0.005 for ischemic heart disease with continuous treatment having a 32% lower risk of all events and a 45% lower risk of ischemic heart disease). In contrast, there was a lower rate of dementia in the intermittent treatment arm when compared with the continuous treatment arm, though these results did not reach statistical significance (HR 1.98, p=0.07 with continuous treatment having a trend toward twice the risk of dementia). Rates of sexual dysfunction, depression, and endocrine outcomes, including osteoporosis and fracture rate, were similar between groups. Several pre-planned sensitivity analyses were performed to assure the validity of these results, and the results were unchanged.
In her discussion, Dr. Hershman discussed why the rate of ischemic events might be higher among men treated with intermittent therapy. She noted that ADT was given a black box warning for having an increased risk of diabetes and cardiovascular disease in 2010 after a consensus meeting of experts. She then described that cycling off and on ADT may cause changes in the coagulation cascade that occur most prominently in the first 6 months after initiation of ADT, causing the risk for intermittent patients to peak during each initiation period. Additionally, changes in estrogen levels at the time of ADT discontinuation may alter coagulation again, increasing risk. Finally, after ADT withdrawal, low testosterone and estrogen may persist, increasing risk.
Next, Dr. Hershman discussed why rates of osteoporosis and fracture might be similar between the intermittent and continuous treatment arms. In general, patients treated with ADT have an increase in risk of vertebral fractures by approximately 40%. All patients in this study received 7 months of ADT/bicalutamide at the start of the study, putting them all at risk for bone loss. Differences in recovery of bone mineral density among men who stop ADT exist. Finally, patients treated on study may have received treatment with bisphosphonates to offset changes in bone mineral density, and this may have made endocrine event rates similar.
Dr. Hershman reviewed that although the study does appear to answer these questions of long-term consequences in a novel and robust way, there are always limitations. She acknowledged the limitations of any claims-based assessment and noted that some claims may have misrepresented actual outcomes. She also reminded us that there was no data on the severity of these complications. The investigators did not adjust for multiple comparisons statistically. Finally, these patients may not perfectly represent patients treated in the “real world” as they were enrolled in a clinical trial and thus had a performance status of 0 or 1 with a lower comorbidity burden than many men with prostate cancer.
This study was novel in that the investigators were able to link patients in a clinical trial with claims-based data, dramatically increasing the ability of the study to follow long-term events. It was uniquely able to demonstrate that men treated with intermittent hormonal therapy developed higher rates of thrombotic and ischemic complications and had a trend toward lower rates of dementia. There also appeared to be no difference in rates of osteoporosis, fracture, or diabetes. The investigators provide a final message to clinicians – “Given the failure of S9346 to prove non-interiority [among men with metastatic prostate cancer], clinicians should be cautious about using IADT [in this patient population].” The strength of this study design and clarify of its outcomes support that cautionary statement.
Presented by Dawn L. Hershman MD, MS at the American Society of Clinical Oncology (ASCO) Annual Meeting - Illumination & Innovation: Transforming Data into Learning - May 29 - June 2, 2015 - Chicago, Illinois USA
Columbia University Medical Center, New York, NY USA
Reported by Alicia K. Morgans, MD, assistant professor of medicine and medical oncologist at Vanderbilt-Ingram Cancer Center, and medical writer for UroToday.com