Elderly patients with diffuse large B-cell lymphoma (DLBCL) who have a deficiency of vitamin D and are being treated with combination chemotherapy that includes the monoclonal antibody rituximab have significantly poorer outcomes, a study by German researchers has found.
Although further confirmation is necessary, the study—available online ahead of print in the Journal of Clinical Oncology (10.1200/JCO.2013.53.4537)—showed that raising vitamin D levels prior to treatment increased rituximab-mediated cellular cytotoxicity. Vitamin D deficiency may also affect other antibodies involved in cellular cytotoxicity in the elderly, said the lead author, Michael Pfreundschuh, MD, Professor and Director of the Clinic for Therapy of Tumor, Blood, Immunological, and Rheumatological Diseases at University Hospital of Saarland in Homburg, Germany.
The deficiency is also associated with a poor response to rituximab in other cancers, notably colorectal and some breast cancers, he noted. “Patients with vitamin D levels less than 8 ng/ml had a significantly worse outcome if they received rituximab. Conversely, the addition of rituximab improved three-year event-free survival by only 16 percent compared with 31 percent in patients with better vitamin D levels.”
The study (first author is Jörg Thomas Bittenbring, MD) involved a subset of patients enrolled in the CHOP-14 trial, which tested a two-week chemotherapy regimen of cyclophosphamide, daunorubicin, vincristine, and prednisone. The researchers also tested six versus eight cycles of biweekly CHOP-14 with or without rituximab in the subsequent RICOVER-60 trial.
The new data are from a further investigation, called RICOVER-noRTh, in which patients received six cycles of cyclophosphamide, doxorubicin, vincristine, and prednisone at an interval of two weeks, plus two cycles of rituximab, with pretreatment evaluation of vitamin D.
“We substituted vitamin D according to the formula suggested in a 2010 German study [E J Endocrinol 2010;162:805-811], with a maximum daily intake of 200,000 units per day, and the target level is usually achieved within two to four days,” he said.
“In our study, elderly DLBCL patients started two years before we realized the impact of vitamin D on treatment results. Thus, the first 250 patients in this study did not receive vitamin D. Starting with 301 patients, all received vitamin D substitution with the above mentioned target level.
We now have 400 patients in the trial, and it will take probably two more years until we know whether vitamin D substitution improves outcome in patients who are severely vitamin D deficient.”
According to the research, a chemoluminescent immunoassay was used to measure serum levels of 25-hydroxyvitamin D3 (25[OH]D3) in 359 patients in the RICOVER 60 trial and 63 subjects in RICOVER-noRTh. Rituximab-mediated cellular cytotoxicity (RMCC) was assessed by a lactate dehydrogenase release assay.
Patients with vitamin D levels of less than 8 ng/mL were found to have an event-free survival rate of 59 percent, and a three-year overall survival rate of 70 percent, while the rates in those with levels above 8 ng/mL were 79 and 80 percent, respectively.
The risk was double for event-free survival in a multivariable analysis adjusting for International Prognostic Index risk factors, but less so for overall survival, and event-free survival was not significantly different in patients with vitamin D levels below or above 8 ng/mL among subjects not treated with rituximab.
An independent evaluation showed that in 63 RICOVER-noRTh patients, rituximab-mediated cell toxicity significantly increased in seven of seven vitamin D-deficient (VDD) patients after their vitamin D levels were normalized.
“VDD impairs RMCC and is therefore a risk factor in elderly patients treated with R-CHOP. That vitamin D substitution improved RMCC strongly suggests that vitamin D substitution enhances rituximab efficacy,” Pfreundschuh said, noting, though, that this remains to be confirmed in prospective trials addressing vitamin D deficiency and substitution—“not only in DLBCL, but also in malignancies treated with other antibodies, of which the major mechanism of action is antibody-dependent cellular cytotoxicity, such as trastuzumab in breast cancer and cetuximab in colorectal cancer.”
The study did have some limitations, however, the researchers noted: Pretreatment serum samples were not mandatory in either the training cohort (RICOVER-60) or the validation cohort in RICOVER-noRTh, and sera were available from only about 30 percent of all patients. Nonetheless, patients with pretreatment sera were representative of the entire RICOVER-60 population.
The optimal vitamin D cutoff point was 8 ng/mL, and the study was restricted to patients older than age 60, who are known to have lower vitamin D serum levels compared with younger patients.
In contrast to the situation in the United States, no 25(OH)D3 is added to milk in Germany, so the lower 25(OH)D3 levels measured might be a result of the chemoluminescent immunoassay used in the study and lower values than with the liquid chromatography-tandem mass spectrometry used in a 2010 study of vitamin D deficiency by researchers at Mayo Clinic (JCO 2010;28:4191-4198).
Pfreundschuh said that he believes vitamin D screening needs to become more widely used in all elderly patients, not just those with this cancer, and that elderly DLBCL patients are not routinely tested.
Multiple Antibodies in Cytotoxic Response
Asked for his opinion for this article, Jonathan W. Friedberg, MD, Professor and Director of the University of Rochester Cancer Center, said the study is very interesting, “not only because the investigators demonstrated that vitamin D deficiency could be a risk factor in such patients and could serve as a prognostic indicator of response to treatment, but also because it suggests that this response may involve multiple antibodies in cytotoxic response.
“This opens the door for future research including vitamin D-positive involvement in antibodies,” he said.
He noted that in the U.S., testing patients for vitamin D deficiency has become much more common, especially due to its role in bone health. “We would expect testing to be higher in cancer patients, and we have looked at it in lymphoma, but I believe vitamin D supplementation is still deficient.”
One reason for this is that there is the potential for toxicity with higher levels of supplementation. But he said there are other questions that need to be resolved: “One big question is whether vitamin D deficiency is a result of patients' illness or whether it contributes to worse outcomes, and whether intervention can correct outcomes. I think the research might support a prospective study of supplementation. But even with such a trial there is the ethical question about keeping supplementation away from patients in a randomized study.”
Matthew Drake, MD, PhD, an endocrinologist at Mayo Clinic, the first author of the 2010 study mentioned above, said that checking 25(OH)D levels in such patients should be routine: “This is a simple blood test, easy to interpret, and can be readily acted upon with replacement, which is inexpensive and easy to administer,” he said. “The potential reasons why this is not done could include that prescribing physicians and/or patients are unaware or disbelieving of this relationship.”
His study, a collaborative trial of 374 newly diagnosed DLBCL patients, found that 50 percent had deficient vitamin D levels based on the commonly used clinical value of a total serum 25(OH)D level of less than 25 ng/mL. Patients with deficient vitamin D levels had a 1.5-fold greater risk of disease progression and a twofold greater risk of dying, compared with patients with optimal vitamin D levels after accounting for other patient factors associated with worse outcomes.
Commenting on the new study's limitations, he said he had concerns that the 8 ng/ml vitamin D level used as the cut-off is very low—much lower than the Institute of Medicine or the Endocrine Society have recommended: “So this would still be in the deficient range,” he said. “Also, the study is retrospective, and so all potential confounding variables may not be included and accounted for. In addition, only eight subjects, seven of whom were evaluated, were included, which should be expanded in future analyses.”
He noted that conducting a large-scale prospective study in which patients with DLBCL are randomized to receive either vitamin D or placebo, with relevant clinical endpoints evaluated, is needed. “However, given these positive findings and those of earlier investigations, including our previous studies here at Mayo in DLBCL and chronic lymphocytic leukemia, recruitment may be challenging given that patients can readily obtain vitamin D supplementation over the counter.”