From Medscape News Today – Kate Johnson Dec 10, 2012
SAN ANTONIO, Texas — Breast cancer patients with insufficient vitamin D levels have a worse prognosis after standard treatment plus zoledronic acid than those with normal levels of vitamin D, according to a preliminary biomarker analysis of the AZURE trial reported here at the 35th Annual San Antonio Breast Cancer Symposium.
“Women who have sufficient vitamin D levels appear to have a much better prognosis,” said lead researcher Robert Coleman, MD, from the University of Sheffield in the United Kingdom. “We should be measuring vitamin D and replenishing it appropriately…. [But] “whether vitamin D as an intervention will change outcome, I don’t know,” he added.
The AZURE trial ( N Engl J Med. 2011;365:1396-1405), published last year by Dr. Coleman’s group, compared standard systemic local therapy given either alone or alongside zoledronic acid 4 mg in 3360 patients with stage II to III breast cancer.
Zoledronic acid is indicated for reducing and delaying bone complications in cancer patients with bone metastases.
After a median follow-up of 53.3 months, the researchers found no difference between the treatment groups for the primary outcome of disease-free survival (hazard ratio [HR] for zoledronic acid, 1.15). However, a preplanned subgroup analysis showed significant benefits in postmenopausal women (HR, 0.75), Dr. Coleman reported.
The researchers undertook the biomarker analysis because “we felt it was appropriate to try to understand some of the biology that might be going on to explain this somewhat unexpected result,” he said.
In a subset of 872 patients from the original study, the researchers looked at baseline serum levels of 25-hydroxyvitamin D, N-terminal propeptide type I procollagen (PINP) — a bone-formation marker — and beta C-terminal telopeptide type I collagen (β-CTx) — a bone resorption marker — to see if they correlated with outcomes.
A PINP level above 70 ng/mL was considered abnormal, as was a β-CTx level below 0.299 ng/mL. For vitamin D, levels under 30 ng/mL were considered insufficient.
Almost half of the women in the subset (46.9%) were premenopausal, 30.5% were more than 5 years postmenopausal, and 14.1% were less than 5 years postmenopausal.
The analysis found that neither PINP nor β-CTx levels were predictive of outcome, “but for vitamin D, there were surprising differences,” said Dr. Coleman.
Specifically, there were no significant differences between patients with abnormal PINP levels and those with normal levels in either time to bone recurrence (hazard ratio [HR], 1.15; P = .5957) or time to distant recurrence (HR, 0.86; P = .4098).
Similarly, patients with normal and abnormal β-CTx levels had similar times to bone recurrence (HR, 1.43; P = .592) and times to distant recurrence (HR, 1.21; P = .2559).
“PINP doesn’t appear to have any prognostic value for predicting either bone or distant recurrence. There’s a trend suggesting that high CTx is associated with worse prognosis, but this is certainly not statistically significant for either bone recurrence or distant recurrence,” he concluded.
In contrast, sufficient vitamin D levels significantly predicted lower risk for bone recurrence (HR, 0.11; P = .0257). There was also a trend toward improved prognosis for distant recurrence (HR, 0.56; 95% confidence interval, 0.31; P = .0519).
Finally, postmenopausal women with sufficient vitamin D levels appear to gain more from zoledronic acid (HR, 0.09) than those with insufficient levels (HR, 0.72); that difference approached statistical significance ( P = .0747).
Premenopausal women responded less well to zoledronic acid. Whether they had sufficient (HR, 1.45) or insufficient (HR, 1.44) levels of vitamin D appeared to have no impact ( P = .9886).
Although the impact of vitamin D levels is clear from the study, “it is not clear that the results provide a signal to supplement,” said Rowan Chlebowski, MD, from the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center in Torrance, California, after the session.
“The Women’s Health Initiative and other studies have shown that only 20% of the difference in vitamin D levels between individuals can be explained by diet, supplements, sunlight exposure, exercise, and body mass index,” Dr. Chlebowski told Medscape Medical News.
“Probably 80% of the difference between individuals is genetically determined. There aren’t any interventions where supplementing deficiency to a target level shows a change of outcome, aside from rickets. So there is a caution in terms of knowing what the immediate clinical implications of the findings are,” he noted.
The AZURE trial was funded by Novartis Pharmaceuticals and the National Cancer Research Network. Dr. Coleman reports being a consultant for Novartis and giving expert testimony for the company. Dr. Chlebowski reports being a consultant/advisor for AstraZeneca, Novartis, and Genentech; being a committee member and meeting participant/lecturer for Celgene, Novartis, and AstraZeneca; and doing scientific study/trial for Amgen. Study coauthor Dr. Brown reports receiving grant/research support from Amgen and Novartis; being on the speaker’s bureau for BMS, Amgen, and Novartis; and being a consultant for Amgen. Coauthor Dr. Cameron reports being a consultant for Novartis.
35th Annual San Antonio Breast Cancer Symposium (SABCS): Abstract S3-4. Presented December 7, 2012.