|Home | About | Journals | Submit | Contact Us | Français|
Obesity has implications for chemotherapy dosing and selection of patients for therapy. Autologous hematopoietic stem cell transplant (AutoHCT) improves outcomes for patients with multiple myeloma, but optimal chemotherapy dosing for obese patients is poorly defined. We analyzed the outcomes of 1087 recipients of AutoHCT for myeloma reported to the CIBMTR between 1995 and 2003 receiving high-dose melphalan conditioning, with or without total body irradiation (TBI). We categorized patients by body mass index (BMI) as normal, overweight, obese, or severely obese. There was no overall effect of BMI on progression-free survival (PFS), overall survival (OS), progression, or non-relapse mortality (NRM). In patients receiving melphalan and TBI conditioning, obese and severely obese patients had superior PFS and OS compared with normal and overweight patients, but the clinical significance of this finding is unclear. More obese patients were more likely to receive a reduced dose of melphalan, but there was no evidence that melphalan or TBI dosing variability affected PFS.
Therefore, current common strategies of dosing melphalan do not impair outcomes for obese patients, and obesity should not exclude patients from consideration of autologous transplantation. Further research is necessary to optimize dosing of both chemotherapy and radiation in obese patients.
Novel anti-myeloma agents have improved outcomes for patients, but the first major advance in myeloma therapy since the 1960s was the demonstration that autologous hematopoietic stem cell transplantation (AutoHCT) improves survival(1, 2). AutoHCT involves dose escalation of chemotherapy beyond the myeloablative threshold to a maximum-tolerated dose defined by other toxicities. Although multiple myeloma (MM) is the commonest indication for AutoHCT(3), the procedure is not curative. Outcomes remain suboptimal, with significant variability in disease control and treatment related toxicity.
Obesity is a potential source of variability in treatment outcomes. Obesity may affect outcomes through alterations in chemotherapy dosing and pharmacokinetics;(4) association with co-morbidities or more aggressive disease, worse stage at diagnosis, and poorer response to chemotherapy;(5, 6) or conversely as a marker for the absence of cancer cachexia. Because of these overlapping and potentially contradictory effects, understanding the effect of obesity on cancer treatment requires careful study of specific situations and appropriate control for confounding factors. The effect of obesity on outcomes from chemotherapy has been studied in other patient populations,(5, 7-17) including patients undergoing hematopoietic stem cell transplantation,(18-25) but not in myeloma patients undergoing AutoHCT.
Conditioning therapy prior to AutoHCT in MM usually involves a single chemotherapeutic agent, melphalan, given near its maximum tolerated dose, which is based on body surface area (BSA). Although the dose of melphalan is directly related to its toxicity and anti-neoplastic efficacy, no data exist to guide dose adjustments in obese patients receiving high-dose melphalan. The effects of obesity on dosing changes and their impact on post transplant outcomes are important for guiding treatment decisions and further research.
The study population included adult recipients of autologous peripheral blood hematopoietic stem cell transplant for MM between 1995 and 2003 reported to the Center for International Blood and Marrow Transplant Research (CIBMTR). The study population was limited to patients receiving a melphalan-based conditioning regimen with or without total body irradiation (TBI), and to transplants done as part of initial therapy, defined as an interval between diagnosis and transplant of ≤18 months. Planned tandem AutoHCT were excluded.
Obesity was defined according to BMI at the time of transplant. BMI was calculated as weight (in kg)/height (in m)2 and categorized using definitions of the WHO and National Heart, Lung, and Blood Institute as: normal weight (18.5≤BMI<25), overweight (25≤BMI<30), obese (30≤BMI<35), and severely obese (BMI≥35)(25). Underweight (BMI<18.5) patients were excluded from the analysis due to the small sample size (n=9) and the likely complicating influence of cancer cachexia due to severe disease. Melphalan doses were reported to the CIBMTR as total mg administered, without specific information as to the treating physician’s intended dose per m2 of BSA or dose modifications for any patient characteristics. We therefore also expressed doses of melphalan as mg/m2 based on BSA calculated from the height and the actual body weight (ABW) at time of transplant, using the formula developed by DuBois and DuBois.(26, 27) Alternative BSA calculations were also performed using ideal body weight (IBW; calculated as [height in m]2 × 22)(27) and adjusted IBW (calculated as IBW + 0.25 × [ABW–IBW]).
Overall survival from transplant was defined as the time from date of transplant to date of death, with survivors censored at the time of last contact. Non-relapse mortality (NRM) was defined as death occurring in the absence of relapse/progressive disease and summarized by the cumulative incidence estimate with relapse/progression as the competing risk. Relapse/progression was defined as the time to first evidence of progression of myeloma according to the standard EBMT/IBMTR criteria(28, 29) and summarized by the cumulative incidence estimate with NRM as the competing risk(29). Progression-free survival (PFS) was defined as survival without progressive disease or relapse from complete response.
Patient-, disease-, and transplant-related variables were described and compared among the BMI groups using the chi-square statistic for categorical variables and the Kruskal-Wallis test for continuous variables. Univariate probabilities of PFS and OS were estimated using the Kaplan-Meier method with the log rank test used for univariate comparisons.(30) Univariate probabilities of NRM and relapse/progression were calculated using cumulative incidence curves to accommodate competing risks.(31)
The hazard ratios of the main outcome of interest (BMI subgroups) and other risk factors potentially associated with relapse/progression, treatment failure and survival were modeled using Cox proportional hazards regression.(30, 32) A stepwise forward/backward model selection approach was used to identify significant risk factors, with the main effect for BMI forced into the model at each step and factors significant at a 5% level kept in the final model. Potential interactions between the main effect and all significant risk factors were tested at the significance level of 0.015, by which criterion a significant interaction was found between TBI and BMI group for progression and PFS and borderline (p=0.012) for OS. Final models for progression, PFS and OS were therefore constructed including this interaction. Adjusted probabilities of PFS and OS were generated from the final Cox models stratified on BMI group and weighted averages of covariate values using the pooled sample proportion as the weight function and were used to estimate likelihood of outcomes in populations with similar prognostic factors. Examination for center effects using a random effects or frailty model found no evidence of a center effect. We compared the factors associated with melphalan dose reduction using logistic regression models and the effect of dose reduction on progression-free survival using Cox proportional hazard models.
Statistical analyses were performed using the statistical package of SAS version 9. All p-values are two-sided and reported without adjustment for multiple comparisons.
The 1,087 patients with MM who met study inclusion criteria included 292 of normal weight, 472 overweight, 198 obese, and 125 severely obese. Median follow-up of survivors is 63 (range 1-144) months, 61 (3-144) months, 60 (3-133) months and 59 (3-131) months for the respective BMI categories.
Baseline patient characteristics are shown in Table 1. Obese patients were younger than their non-obese counterparts. There was a higher prevalence of diabetes among obese patients. Obese transplant recipients also had less frequent hypercalcemia, severe anemia, and renal insufficiency. A higher proportion of severely obese patients were in complete or partial response at time of transplant.
In the study population as a whole, there was no clear effect of BMI on PFS, OS, progression or NRM. We identified a statistically significant interaction between BMI and conditioning regimen for PFS and progression. Among those receiving melphalan alone, there was no clear association between BMI and these outcomes, with PFS at 5 years of 17% of normal weight patients, 18% of overweight patients, 21% of obese patients, and 14% of severely obese patients (p=0.65). Among those receiving melphalan and TBI, obese and severely obese patients had superior PFS and OS than did normal and overweight patients, with PFS at 5 years of 23% in normal weight patients, 17% in overweight patients, 43% in obese patients, and 55% in severely obese patients (p=0.005). P-values for the interactions between BMI and conditioning regimen were highly significant for PFS (p=0.0063) and progression (p=0.0085), with borderline significance for OS (p=0.012), but not significant for NRM (p=0.43). The effect of BMI on outcomes after transplant was therefore restricted to patients receiving melphalan with TBI, with obese patients having better outcomes, mediated primarily through a lower risk of disease progression.
To exclude confounding by baseline imbalances, we constructed multivariable models of PFS, OS, and progression using all potential confounders. These models included each variable individually, in groups of related variables, and in a single model using all variables, none of which showed any evidence of confounding, with no change in the estimate of effect of BMI on outcomes for patients receiving melphalan alone or melphalan with TBI (data not shown). Final multivariable models, constructed using a forward/backward stepwise algorithm, confirmed the effects of BMI on PFS, progression and OS, as shown in Tables Tables2,2, ,3,3, and and4.4. A multivariate model for NRM (not shown), showed no evidence of confounding, and BMI did not have any clear effect on this outcome. Estimated probabilities of all these outcomes based on the final multivariate models are shown in Figure 1.
Doses of melphalan were compared among BMI groups as absolute doses and as doses per m2 of BSA calculated using ABW, IBW, and adjusted IBW. As expected with BSA-based dosing, patients who were more obese received higher absolute doses of melphalan, both when melphalan was given alone and when it was given in combination with TBI (Figure 2). With both conditioning regimens, patients who were more obese received a lower dose per m2 of actual BSA (i.e. using ABW to calculate BSA). Using IBW or adjusted IBW to calculate BSA resulted in normalized doses that were closer among BMI groups, though there were still significant differences among the groups. When compared on the basis of melphalan dose per kg of body weight,(33) more obese patients received a lower dose of melphalan per kg of ABW.
We further investigated the effect of chemotherapy dosing decisions in obese patients. A full dose of melphalan was defined as 200 mg/m2 (calculated by ABW) for conditioning regimens using melphalan alone and as 140 mg/m2 for transplants using melphalan with TBI. A reduced dose of melphalan was defined as <90% of the full dose. Using this definition, reduced doses of melphalan were given to 78% of severely obese, 56% of obese, 32% of overweight, and 11% of normal weight patients (p<0.0001). Therefore, the odds of dose reduction were 30 (95% confidence interval 17-53) times higher for severely obese patients, 11 (95%CI 7-17) times higher for obese patients, and 4 (95%CI 3-6) times higher for overweight patients, compared with normal weight patients. Dose reduction was also more common for patients with renal insufficiency (odds ratio 2.0, 95%CI 1.2-3.4), history of hypertension (OR 1.7, 95%CI 1.2-2.4) or diabetes (OR 1.8, 95%CI 1.2-2.7), non-Caucasian race (OR 1.4, 95%CI 1.1-2.0), or transplant in the period from 2001-2003 (OR 1.4 compared to transplant in 1995-2000, 95%CI 1.1-1.8). Dose reduction was not associated with other variables, including age, performance status, number of prior therapies, or disease status at time of transplant.
There was no evidence of an effect of melphalan dose reduction on PFS. Receipt of a reduced dose of melphalan was associated with a univariate hazard ratio for PFS of 1.07 (95%CI 0.92-1.23), and in a multivariate analysis (controlling for renal function, performance status, age, race, gender, history of hypertension or diabetes, disease status, and year of transplant) the hazard ratio associated with melphalan dose reduction was 0.88 (95% CI 0.70-1.10). We also found no effect of melphalan dose on PFS, regardless of whether the dose was specified as total melphalan administered, as the dose per m2 of body surface area (with, in successive analyses, the BSA calculated using ABW, adjusted IBW, or IBW), or as the dose per kg of ABW. For the absolute dose of melphalan, the hazard ratio associated with a 10 mg increase in dose was 0.99 (95%CI 0.98-1.01); the hazard ratios per 10 mg/m2 increase in dose were 1.00 (95%CI 0.97-1.04), 0.99 (95%CI 0.96-1.02), and 1.01 (95%CI 0.99-1.04) when BSA was calculated with actual, adjusted ideal, and IBW, respectively; and the hazard ratio per 1 mg/kg increase in dose was 1.05 (95%CI 0.95-1.16) when the dose was calculated per kg of ABW. Repeating these analyses in the population of obese and severely obese patients yielded similar results (data not shown). Therefore, none of these analyses showed any effect of variation in chemotherapy dose on PFS.
We further investigated the reason for the restriction of the effect of obesity to TBI-containing transplants. Most patients received a planned dose of 12 Gy (65% of normal and overweight, 74% of obese, and 68% of severely obese patients), with a few patients receiving 13 Gy (9% of normal and 10% of overweight, none of the obese or severely obese), 10-11 Gy (12% of normal, 13% of overweight, 16% of obese, and 23% of severely obese), or <10 Gy; these differences were not significant (overall p-value 0.75). There was no discernible effect of TBI dose on PFS and no evidence that the TBI dose confounded the effect of obesity on PFS (data not shown).
We have analyzed a large cohort of patients receiving high-dose melphalan based AutoHCT for initial therapy of myeloma. We did not find any differences in outcome between obese and non-obese patients receiving a conditioning regimen of single-agent melphalan. This finding suggests that obese patients can receive high-dose melphalan AutoHCT without increased treatment-related mortality and that obesity should not be a contraindication to undergoing autologous transplantation. Participating transplant centers reported total melphalan doses administered but not information regarding the specific intention of treating physicians in adjusting melphalan doses for individual patients. We therefore do not know which patients received reduced doses based on ideal or adjusted ideal body weight (or with calculated BSA capped to define a maximum allowable dose). Previously published literature suggests that variation in approach to dose adjustment is substantial,(34, 35) and a variety of dosing strategies were likely used in the patients in our study. This is clear in the significant minority of obese patients in our cohort who received doses within 10% of a full dose calculated according to actual body weight (44% of obese patients and 22% of severely obese patients). Our results suggest that current strategies used in clinical practice for adjusting melphalan doses in obese patients (i.e. calculating doses based on ideal or adjusted ideal body weight for some but not all patients) do not appear to impair outcomes in this population. However, our data also confirm previous findings that dosing strategies in obese patients are variable, with a wide range of melphalan doses. Further research to determine optimal dosing for these patients may be helpful in decreasing variability in toxicity and efficacy.
Our results also show that among patients receiving melphalan with TBI, a higher BMI was associated with improved PFS, OS, and risk of progression. The reason for the restriction of a beneficial effect of obesity to TBI-containing conditioning regimens is unclear. One possible explanation is that the distribution of TBI dose through body tissues differs between obese and thin patients. Standard practice calculates the target radiation dose at the midplane of the body, but deposition of the radiation dose is not uniform, and increasing body size (as measured by anteroposterior distance) requires higher doses delivered to the more superficial tissues to achieve the same midplane dose.(36) Differences in body fat content are generally assumed to have little difference on dose distribution, but it is possible that obese patients’ increased size would result in higher doses to the bone marrow, leading to improved anti-myeloma efficacy. It is also possible that the metabolism of tumor cells in obese patients differs in a way that makes the cancer more susceptible to radiation, leading to a simple effect of increased radiation efficacy in obese patients. We cannot completely exclude confounding as an explanation for the improved outcomes in obese patients receiving TBI, in that obese transplant recipients may have had less aggressive myeloma or less cancer cachexia in ways not captured by our measured covariates. Such differences would not account for a differential effect restricted to TBI conditioning unless selection of the conditioning regimen at some centers depended on a combination of body size and overall health, so that TBI was offered to healthier obese patients. However, we found no evidence of confounding by any characteristics of the patients, their myeloma, or their therapy, although we are limited by the lack of complete data on some of the newer prognostic factors such as International stage or beta-2 microglobulin levels. We therefore conclude that further investigation is warranted to fully understand the effect of obesity in the setting of TBI and determine whether delivering therapeutic doses to marrow-containing structures could be further optimized.
The regimen of melphalan and TBI is no longer commonly used for myeloma due to evidence of similar results and less toxicity with melphalan alone(37). This may limit further investigation into this regimen in the setting of myeloma therapy, but TBI is used in patients with some types of lymphoma and commonly in patients with acute leukemia undergoing allogeneic hematopoietic stem cell transplant. Further research will be necessary to determine whether the association between obesity and outcomes applies to these other patient populations, for whom improved targeting of radiation doses could lead to better treatment outcomes.
Our results differ from previously published studies of AutoHCT, but no other study has examined the same specific population. Two studies (in acute myeloid leukemia or lymphoma) have found obesity to be associated with worse post-transplant outcomes,(21, 24) but analysis of a larger lymphoma cohort from the CIBMTR found no difference in outcomes.(18, 22) Other studies have been limited by the inclusion of multiple diseases, different conditioning regimens, or both autologous and allogeneic transplants, and have found either no effect of obesity(18) or an increased risk of non-relapse mortality.(19) None of these trials examined results in patients with myeloma or examined separately outcomes using TBI-containing regimens, and their results are therefore not directly comparable with the results of our study.
Our study found no evidence that variability in chemotherapy dosing among obese patients leads to differences in outcomes, nor that obese patients are at higher risk of treatment-related mortality or disease progression. In fact, among patients receiving melphalan and TBI, obese patients had a lower risk of relapse. The current commonly used strategy of reducing melphalan doses (calculating based on ideal or adjusted ideal body weight) does not appear to impair outcomes for obese patients, and obesity should not exclude patients from consideration of autologous transplantation. Further research is necessary to optimize dosing of both chemotherapy and radiation in obese patients.
Dr. Vogl is supported by a Special Fellow in Clinical Research Award from the Leukemia & Lymphoma Society, a Scholar Award from the American Society of Hematology, and grant K23CA130074 from the National Cancer Institute.
The CIBMTR is supported by Public Health Service Grant/Cooperative Agreement U24-CA76518 from the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI) and the National Institute of Allergy and Infectious Diseases (NIAID); a Grant/Cooperative Agreement 5U01HL069294 from NHLBI and NCI; a contract HHSH234200637015C with Health Resources and Services Administration (HRSA/DHHS); two Grants N00014-06-1-0704 and N00014-08-1-0058 from the Office of Naval Research; and grants from AABB; Aetna; American Society for Blood and Marrow Transplantation; Amgen, Inc.; Anonymous donation to the Medical College of Wisconsin; Astellas Pharma US, Inc.; Baxter International, Inc.; Bayer HealthCare Pharmaceuticals; Be the Match Foundation; Biogen IDEC; BioMarin Pharmaceutical, Inc.; Biovitrum AB; BloodCenter of Wisconsin; Blue Cross and Blue Shield Association; Bone Marrow Foundation; Buchanan Family Foundation; Canadian Blood and Marrow Transplant Group; CaridianBCT; Celgene Corporation; CellGenix, GmbH; Centers for Disease Control and Prevention; Children’s Leukemia Research Association; ClinImmune Labs; CTI Clinical Trial and Consulting Services; Cubist Pharmaceuticals; Cylex Inc.; CytoTherm; DOR BioPharma, Inc.; Dynal Biotech, an Invitrogen Company; Eisai, Inc.; Enzon Pharmaceuticals, Inc.; European Group for Blood and Marrow Transplantation; Gamida Cell, Ltd.; GE Healthcare; Genentech, Inc.; Genzyme Corporation; Histogenetics, Inc.; HKS Medical Information Systems; Hospira, Inc.; Infectious Diseases Society of America; Kiadis Pharma; Kirin Brewery Co., Ltd.; The Leukemia & Lymphoma Society; Merck & Company; The Medical College of Wisconsin; MGI Pharma, Inc.; Michigan Community Blood Centers; Millennium Pharmaceuticals, Inc.; Miller Pharmacal Group; Milliman USA, Inc.; Miltenyi Biotec, Inc.; National Marrow Donor Program; Nature Publishing Group; New York Blood Center; Novartis Oncology; Oncology Nursing Society; Osiris Therapeutics, Inc.; Otsuka America Pharmaceutical, Inc.; Pall Life Sciences; Pfizer Inc; Saladax Biomedical, Inc.; Schering Corporation; Society for Healthcare Epidemiology of America; Soligenix, Inc.; StemCyte, Inc.; StemSoft Software, Inc.; Sysmex America, Inc.; THERAKOS, Inc.; Thermogenesis Corporation; Vidacare Corporation; Vion Pharmaceuticals, Inc.; ViraCor Laboratories; ViroPharma, Inc.; and Wellpoint, Inc. The views expressed in this article do not reflect the official policy or position of the National Institute of Health, the Department of the Navy, the Department of Defense, or any other agency of the U.S. Government.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.