Portal de Periódicos da CAPES

Impact of body mass index on incidence of febrile neutropenia and treatment‐related mortality in U nited S tates veterans with diffuse large B‐cell lymphoma receiving rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone

2014; Wiley; Volume: 167; Issue: 5 Linguagem: Inglês

10.1111/bjh.13026

1365-2141

Arun Ganti, Weijian Liu, Suhong Luo, Kristen M. Sanfilippo, Ryan Roop, Ryan C. Lynch, Peter A. Riedell, Katiuscia O’Brian, Graham A. Colditz, Kenneth R. Carson,

Chemotherapy-induced cardiotoxicity and mitigation

In 2012, guidelines from the American Society of Clinical Oncology (ASCO) summarized the literature on chemotherapy dosing based upon actual body weight versus ideal or adjusted body weight (Griggs et al, 2012). The conclusion, largely drawn from studies of solid tumor patients, was that chemotherapy administration based upon actual body weight did not increase short-term toxicities. This resulted in the recommendation against the use of empiric dose-reduction solely due to obesity status (Griggs et al, 2012). To better understand if this recommendation could be applied to patients with diffuse large B-cell lymphoma (DLBCL), we examined the associations between body mass index (BMI) and relative dose-intensity with the outcomes of febrile neutropenia and treatment-related mortality in a large cohort of DLBCL patients treated with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone). We identified a retrospective cohort of patients with DLBCL within the United States Veterans Health Administration (VHA) using previously published methods and inclusion and exclusion criteria (Carson et al, 2012). Details of chemotherapy, radiotherapy, supportive care medications and hospitalizations were determined through chart abstraction. Weight and height measurements within 1 month of treatment initiation were used to calculate BMI and body surface area (BSA). Patients were categorized as underweight, normal-weight, overweight or obese in accordance with the World Health Organization (WHO) classification (WHO Expert Committee, 1995). The Romano comorbidity index was calculated using International Classification of Disease (ICD)-9 codes (Romano et al, 1993). Average relative dose-intensity was calculated utilizing methods previously published (Lyman et al, 2004). The expected number of chemotherapy cycles was 3 in stage I/II patients receiving radiotherapy and 6 in all other patients. Chemotherapy doses <85% of expected in the first cycle were considered ‘dose-reduced’. Febrile neutropenia was determined by chart abstraction, and was considered present based upon review of electronic medical records. Treatment-related mortality was defined as death within 30 d of last chemotherapy administration. Logistic regression evaluated variables associated with treatment-related mortality in the overall cohort and then among those without first cycle dose-reductions. Missing data on lactate dehydrogenase (LDH) level, stage, and B-symptoms were handled through use of multiple imputation (Sterne et al, 2009). All statistical analyses were performed using sas version 9.2 (SAS Institute, Cary, NC, USA). All P-values reported are 2-tailed and P-values <0·05 were considered significant. We identified a total cohort of 1241 DLBCL patients who received R-CHOP within the VHA system. Univariate comparisons stratified by BMI are presented in Table 1. There were no significant differences in the average relative dose-intensities or levels of granulocyte colony-stimulating factor (G-CSF) support among the BMI groups. Obese patients were not significantly more likely to undergo first cycle dose-reduction. The incidence of febrile neutropenia and treatment-related mortality among the three groups differed significantly, with obese patients experiencing less of both. Across all treatment cycles, obese patients were significantly less likely to experience febrile neutropenia (obese: 17·8%, overweight: 23·0%, normal-weight: 28·9%, age-adjusted P < 0·001). Obese patients were also significantly less likely to experience treatment-related mortality (obese: 5·5%, overweight: 8·1%, normal-weight: 11·3%, age-adjusted P = 0·017). In multivariate logistic regression controlling for age, stage, LDH level at diagnosis, B-symptoms, comorbidity score and G-CSF use, obesity was associated with decreased treatment-related mortality compared to normal-weight patients [Odds ratio (OR) 0·50; 95% confidence interval (CI) 0·28–0·90]. Overweight patients demonstrated a non-significant trend towards reduced treatment-related mortality (OR 0·73; 95% CI 0·46–1·14). Among patients without first cycle dose-reductions, obesity was still associated with decreased treatment-related mortality (OR 0·46; 95% CI 0·24–0·90), and overweight patients demonstrated a non-significant trend towards reduced treatment-related mortality (OR 0·69; 95% CI 0·42–1·14; Table 2). Consistent with the ASCO guidelines (Griggs et al, 2012), we found no evidence that treatment of overweight and obese DLBCL patients with actual body weight chemotherapy doses increased the incidence of febrile neutropenia or treatment-related mortality. In fact, we observed a significantly decreased incidence of febrile neutropenia among obese patients despite similar average relative dose-intensities and G-CSF support. Furthermore, obesity was associated with decreased treatment-related mortality after controlling for age, G-CSF use and other prognostic factors. This relationship held among patients who received >85% of expected doses in the first treatment cycle, suggesting that fears of excess febrile neutropenia and treatment-related mortality among obese patients are unfounded in this population. This is of particular importance in DLBCL as studies have suggested that achieving average relative dose-intensity >90% is associated with improved long-term outcomes (Bosly et al, 2008). The ASCO guidelines also raised the question of whether dose-reductions might compromise efficacy in obese patients (Griggs et al, 2012). Evidence suggests improved overall survival among patients experiencing neutropenia or leucopenia during chemotherapy (Shitara et al, 2011). Due to the confounding by indication that is inherent in observational studies, we were unable to fully answer this question (Jepsen et al, 2004). While a trial randomizing obese patients to R-CHOP doses based upon actual versus ideal body weight would provide a clear answer, it is unlikely that such a study will ever be performed. The strengths and limitations of this study should be highlighted. As the VHA is the largest integrated health system in the United States, we were able to assemble a large patient cohort with detailed patient information. However, consistent with the historic military demographics, this cohort was almost entirely men, which may limit the applicability of our findings. An additional limitation of our study is that we could not quantify differences in long-term toxicities among BMI groups. Doxorubicin is known to cause a significant, dose-dependent incidence of heart failure (Gharib & Burnett, 2002). The effect of doxorubicin dosing, based upon actual body weight, on the incidence of doxorubicin-induced cardiomyopathy remains unknown. Overall, this study supports the application of the ASCO obesity guidelines in overweight and obese DLBCL patients (Griggs et al, 2012). Specifically, we found no evidence that treatment with full weight-based therapy in overweight or obese patients increases the risk of febrile neutropenia or treatment-related mortality. Until data emerges suggesting otherwise, in the absence of other contraindications, oncologists treating overweight and obese DLBCL patients with R-CHOP should use actual body weight in dosing calculations. Conception and design: Arun Ganti, Graham A. Colditz, Kenneth R. Carson. Financial support: Arun Ganti, Graham A. Colditz, Kenneth R. Carson. Administrative support: Katiuscia O'Brian. Collection and assembly of data: Arun Ganti, Weijian Liu, Suhong Luo, Kristen M. Sanfilippo, Ryan Roop, Ryan Lynch, Peter Riedell, Katiuscia O'Brian. Data analysis and interpretation: Arun Ganti, Weijian Liu, Suhong Luo, Kristen M. Sanfilippo, Ryan Roop, Ryan Lynch, Peter Riedell, Graham A. Coldtiz, Kenneth R. Carson. Manuscript writing: Arun Ganti, Kristen M. Sanfilippo, Graham A. Coldtiz, Kenneth R. Carson. Final approval of manuscript: All authors. The authors have no competing interests to disclose. This study was funded in part by: The Barnes-Jewish Hospital Foundation, The American Cancer Society grant IRG 58-010-52, the American Society of Hematology Trainee Research Award, and the National Cancer Institute at the National Institutes of Health grants U54CA155496 and KM1CA15608. The content is solely the responsibility of the authors and does not necessarily represent the official views of: The United States Department of Veterans Affairs, Washington University, Saint Louis University, Barnes-Jewish Hospital, the American Cancer Society, the National Cancer Institute or the National Institutes of Health, or the American Society of Hematology.