Portal de Periódicos da CAPES

Autologous stem cell transplant is feasible in very elderly patients with lymphoma and limited comorbidity

2011; Wiley; Volume: 87; Issue: 4 Linguagem: Inglês

10.1002/ajh.23108

1096-8652

Rebecca Elstrom, Peter Martin, Sandra M. Hurtado Rúa, Tsiporah B. Shore, Richard R. Furman, Jia Ruan, Roger Pearse, Morton Coleman, Tomer M. Mark, John P. Leonard, Usama Gergis,

Immune Cell Function and Interaction

In patients with recurrent Hodgkin or non-Hodgkin's lymphoma, autologous stem cell transplantation (ASCT) can offer potential for cure or long-term remission. Because of potential toxicity, elderly patients are usually not considered candidates, but data regarding tolerability and efficacy in this group are lacking. The transplant database at Weill Cornell Medical College was reviewed to identify patients with lymphoma undergoing ASCT at age 69 or greater. Clinical data and comorbidities were correlated with outcome. Twenty-one patients were identified. Sixteen of 19 evaluable patients (76%) achieved complete remission following ASCT, while 2 patients died before response assessment. Median progression-free survival following ASCT was 8 months and median overall survival was 18 months. Age was not predictive of overall survival, but patients 75 and older had inferior progression-free survival compared to younger patients. High-risk status by hematopoietic stem cell transplant comorbidity index (HCT-CI) was associated with short overall survival and high transplant-related mortality. ASCT is feasible and of potential benefit in selected elderly lymphoma patients. Consideration of comorbidities, rather than age alone, may allow selection of patients likely to tolerate and benefit from ASCT. Non-Hodgkin's lymphoma increases in incidence with advanced age. High dose chemotherapy with autologous stem cell transplantation (ASCT) plays an important therapeutic role, particularly in patients with chemosensitive relapses of curable histologies such as diffuse large B cell (DLBCL) and Hodgkin lymphoma (HL), but many patients have not been considered eligible for this approach due to advanced age. The question of whether age alone should be an exclusionary factor for ASCT is controversial. Recent single center retrospective studies reported conflicting results on the impact of age on transplant outcomes. While limited data demonstrating its feasibility in patients >60 years of age are available, most of the patients included in these studies are between the ages of 60 and 70. Although some reports have suggested higher than expected nonrelapse mortality (NRM) in patients over 60 years of age [1], this finding is not consistent [2-5]. Comorbidity refers to a medical condition existing simultaneously with but independently of a primary disorder. The Charlson comorbidity index [6] (CCI) was developed as a tool to predict mortality in hospitalized patients, assigning a weight to a variety of disorders such as diabetes mellitus, liver dysfunction or pulmonary disease, together defining a score which is predictive of mortality. The hematopoietic stem cell transplant comorbidity index (HCT-CI) [7] represents a modification of the CCI to better represent comorbidities identified in the allogeneic stem cell transplant population. Major differences between these indices include incorporation of objective data for measurement of pulmonary, renal, and hepatic dysfunction, in addition to symptoms, and consideration of any prior malignancy as a risk factor, not just malignancies initially treated within 5 years of transplantation. The presence of significant comorbidities, as measured by the CCI or HCT-CI, predisposes to increased toxicity in patients over 60, but this factor has been inconsistently associated with mortality. Many patients in the eighth decade of life or beyond have excellent performance status and prelymphoma overall health. At our center, age has not been routinely used as an exclusion criterion for this treatment modality. We therefore examined outcomes of patients who were 69 and older at the time of transplantation to determine whether ASCT is feasible in this population, and whether comorbidities are predictive of NRM and overall survival (OS). Twenty-one patients aged 69 or greater (range 69–86, median 71) were identified who had undergone autologous stem cell transplant for lymphoma and had records available for review; characteristics are shown in Table I. Five patients were 75 years of age or older, and 2 patients were over the age of 80 (83 and 86 years old). All patients underwent cytotoxic conditioning regimens and subsequently received autologous stem cells. Conditioning regimens for most patients consisted of BCNU, cyclophosphamide, and etoposide (BCV), BCNU, etoposide, cytarabine, and melphalan (BEAM), or similar regimens. Two patients were treated with cyclophosphamide and total body irradiation (TBI), and one received mitoxantrone, thiotepa, and cyclophosphamide. Neutrophil engraftment occurred in 19 patients, while 2 patients died before engraftment could be evaluated. Platelet engraftment occurred in 16 patients, while 3 remained platelet-transfusion dependent. Median time to neutrophil engraftment was 11 days (range 9–26 days), and median time to platelet engraftment was 21 days (range 13–51 days). Sixteen patients (76%) achieved a complete remission (CR) following ASCT, while three patients did not achieve CR. Two patients (10%), ages 73 and 75, died of infection before engraftment and were unable to undergo response assessment. Two more patients, ages 70 and 73, died within or shortly after 100 days of treatment-related complications (infection and cardiac arrhythmia), for a total of 4 (19%) early treatment-related deaths (TRM). At median follow-up of 25 months (range 7–48 months), median progression-free survival (PFS) was 8 months (Fig. 1A) and median overall survival (OS) was 18 months (Fig. 1B). Ten patients (48%) survived in remission for at least 1 year following transplant. (A) Progression-free survival for all patients. (B) Overall survival for all patients. Eight patients were alive at last follow-up, of whom 7 remained in remission, while 13 patients had died. Of the eight deaths that occurred within the first year, four patients died of TRM and four died due to progressive lymphoma. Five deaths occurred more than 1 year after transplant: two patients died of progressive lymphoma, one died of unknown causes, one died of secondary AML at 3 years, and one died of intracranial hemorrhage 4 years following transplantation. Progression-free survival was superior in younger patients when compared to those 75 and older (P = 0.004). However, although there was a trend to better overall survival in younger patients, this trend did not reach statistical significance (P = 0.084). Conditioning regimen was not predictive of either PFS or OS. There was no association between CD34+ cell dose, time to neutrophil engraftment, or time to platelet engraftment and either PFS or OS. HCT-CI and CCI were calculated for all patients (Table II). Using the HCT-CI, three patients were of high risk, eight patients were intermediate risk, and ten patients were low risk. Using the CCI, one patient was of high risk, seven had intermediate risk, and thirteen were low risk. Considering the four patients who died within 100 days of TRM, three were high risk by HCT-CI, and one was low risk, whereas using the CCI, one patient was considered high risk, two intermediate risks, and one low risk. Although there was no difference in OS between low and intermediate risk patients by either index, patients determined to be high risk by HCT-CI experienced extremely poor survival, as shown in Fig. 2 (median 0.3 months vs. 19 months for nonhigh-risk patients, P = 0.000005). Only one patient was identified to be high risk by CCI, limiting the predictive utility of this index. Kaplan–Meier estimate of overall survival by HCT-CI (risk factor—1: low and intermediate versus 2: high risk). Patients of high risk by HCT-CI have inferior overall survival when compared to those of low or intermediate risk (P = 0.000005). In summary, we sought to assess the feasibility of ASCT for lymphoma in a group of very elderly patients and to evaluate the utility of pretransplant validated comorbidity assessments to predict outcome. Our data show that autologous stem cell transplantation is feasible and of potential benefit in selected very elderly patients with lymphoma. Others have studied the impact of age on toxicity and mortality in the setting of stem cell transplantation. Although some have reported increased toxicity and/or TRM in patients over 60 as compared to younger patients [1, 8-10], others have not observed this difference [2-5]. Two previous reports have evaluated the impact of comorbidities on outcome in older patients undergoing ASCT. Wildes et al. [2] compared 59 patients aged 60 and older to younger patients, showing no difference overall in DFS or OS, but a significant impact of CCI on TRM and OS. Hosing et al. [3] found that, in patients over age 65 undergoing ASCT, HCT-CI predicted for increased toxicity during transplant, but had no impact on OS. Andorsky et al. recently published the first data of which we are aware evaluating ASCT in lymphoma patients 70 years of age and older [11]. In this small group of 17 patients, 1-year nonrelapse mortality was 35%. CCI and HCT-CI were calculated but were not predictive for survival in this group. No pretransplant characteristics were identified that could help identify patients who would be at prohibitively high risk of this procedure. HCT-CI was predictive of survival in the group of patients presented here, supporting previous reports in younger patients that comorbidities are likely a more important factor in determining potential benefit of transplant than age. Although it is not clear why our results differ from those of Andorsky et al., it is possible that it relates to either the very small sample size or lack of inclusion of higher-risk patients in the Andorsky report. In our small cohort, the HCT-CI was more sensitive than the CCI in capturing high-risk patients. In fact, 3 out of 4 (75%) patients who died within 100 days of transplant were identified as high risk by the HCT-CI. This could be the basis for testing the validity of the HCT-CI in ASCT in elderly patients, and possibly across all age groups. Within this elderly population, when patients 75 and older were compared to those <74 years of age, lymphoma-specific outcomes appeared inferior. This finding suggests that, although ASCT appears to be tolerable in the setting of limited comorbidity, it may be less effective therapy for this older group of patients. Given the limited numbers and lack of statistical power, firm conclusions regarding the impact of age alone will require a larger sample size. We recognize several limitations to our study, including the retrospective nature of the data collection with potential associated selection bias, the fact that it is a single institution experience, and the small sample size, limiting firm conclusions regarding association of clinical factors with outcome. Notwithstanding these shortcomings, our findings extend ASCT as a feasible option in a selected group of very elderly patients. The paucity of published data addressing aggressive treatment options in this age group calls for more effort to discern patients who can benefit from this approach. In conclusion, age alone need not be an exclusionary factor for ASCT in elderly patients with lymphoma. Selected patients with limited comorbidities can benefit from this treatment modality in spite of advanced age. We identified all patients age 69 or greater with a diagnosis of Hodgkin lymphoma (HL) or non-Hodgkin's lymphoma (NHL) who underwent ASCT between June 2003 and December 2009. Patients were identified by review of the stem cell transplant database at Weill Cornell Medical College. Age, date of transplant, conditioning regimen, engraftment data (neutrophils and platelets), response, and survival data were derived from review of medical records. Determination of candidacy for ASCT and conditioning regimen were at the discretion of the treating physician. Baseline Charlson comorbidity risk index (CCI) and hematopoietic stem cell transplant comorbidity index (HCT-CI) were calculated for all patients. Comorbidity data were collected from review of medical records, and comorbidities were scored as previously published [6, 7]. For both indices, a score of 0 was defined as low risk, a score of 1–2 was defined as intermediate risk, and a score of 3 or higher was considered high risk. Comorbidity risk scores were correlated with NRM, OS, and PFS. Primary endpoints for the analysis were progression-free survival (PFS) and overall survival (OS). OS was defined as the time from transplant until death or last follow up, and PFS was defined as the time from transplant until disease progression, death or last follow up. Treatment-related mortality (TRM) was defined as death within 100 days of transplant not attributed to progression of lymphoma. Analyses of OS and PFS were performed using the Kaplan–Meier method [12]. Confidence intervals were computed using a 95% confidence level. Confidence intervals for the median survival times were computed using Brookmeyer and Crowley method [13]. Analysis was carried out using R—A language and environment for statistical computing [14]. This study was reviewed and approved by the Investigational Review Board of Weill Cornell Medical College. Rebecca L. Elstrom*, Peter Martin*, Sandra Hurtado Rua , Tsiporah B. Shore , Richard R. Furman*, Jia Ruan*, Roger N. Pearse* , Morton Coleman*, Tomer Mark* , John P. Leonard*, Usama Gergis , * Center for Lymphoma and Myeloma, Weill Cornell Medical College, New York, New York;, Department of Public Health, Division of Biostatistics and Epidemiology, Weill Cornell Medical College, New York, New York;, Bone Marrow Transplant Program, Weill Cornell Medical College, New York, New York.