Rigor prophylaxis in stage IV melanoma and renal cell carcinoma patients treated with high dose IL-2

Abstract

Background: Rigors are a significant adverse event during interleukin-2 (IL2) therapy for metastatic melanoma and renal cell carcinoma. Meperidine has been a mainstay for rigor prophylaxis but there is a paucity of data regarding possible alternatives. Methods: Ninety one patients receiving IL2 therapy for metastatic renal cell carcinoma and melanoma at Huntsman Cancer institute (HCI), Utah from May 2009 to October 2016 were retrospectively evaluated for rigor prophylaxis. Forty two patients received meperidine and 49 received tramadol. Rigors were tabulated using the proxy of number of doses of as needed (PRN) rigor medications and normalized by IL2 doses. Other outcomes of fever, hypotension, and renal insufficiency were noted on a binary scale and normalized by cycles. Statistical analysis was performed utilizing univariate and multivariate negative binomial models. Results: Ninety one patients were identified with metastatic melanoma or RCC who received high dose IL2 therapy. Forty two received meperidine and 49 received tramadol prophylaxis for rigors. Univariate negative binomial analysis shows incidence rate ratios (IRR): fever 0.41 (95% CI 0.28-0.62, p-value < 0.001), hypotension 1.7 (95% CI 1.11-2.61, p-value 0.015), renal insufficiency 0.58 (95% CI 0.35-0.98, p-value 0.041), rigors per all PRN meds 1.01 (95% CI 0.79-1.28, p-value 0.964), and rigors via opioid PRN meds 0.85 (95% CI 0.67-1.07, p-value 0.168). Multivariate negative binomial analysis shows IRR: fever 0.59 (95% CI 0.28-1.24, p-value 0.163), hypotension 0.93 (95% CI 0.43-2.03, p-value 0.864), renal insufficiency 1.1 (95% CI 0.52-2.32, p-value 0.807), rigors per al PRN meds 0.92 (95% CI 0.67-1.26, p-value 0.604), and rigors via opioid PRN 0.9 (95% CI 0.65-1.26, p-value 0.554). Conclusion: Univariate models indicated meperidine pre-treatment was associated with significantly lower rates of fever and renal insufficiency whereas tramadol was associated with significantly lower rate of hypotension. However, when controlled for demographics and other treatment differences, these differences were no longer significant.