Favorable Outcomes in Elderly Patients Undergoing High-Dose Therapy and Autologous Stem Cell Transplantation for Non-Hodgkin Lymphoma

Abstract

High-dose therapy and autologous stem cell transplantation (HDT-ASCT) can offer potential long-term remission or cure in patients with non-Hodgkin lymphoma (NHL). Limited experience is available on the safety and efficacy of HDT-ASCT in elderly patients. This is a single-center, retrospective study examining outcomes of HDT-ASCT for 202 NHL patients age 60 years and older between January 2001 and December 2012. Overall survival (OS) and progression-free survival (PFS) were analyzed according to age at HDT-ASCT, hematopoietic cell transplantation comorbidity index (HCT-CI), NHL histology, and remission status at the time of HDT-ASCT. The median age was 65 years (range 60–74) and the majority had either diffuse large B-cell lymphoma (DLBCL, n=73, 37%) or mantle cell lymphoma (MCL, n=69, 34%). One hundred and fifteen patients (57%) had high HCT-CI scores at the time of HDT-ASCT. With a median follow-up of 3.6 years (range 0.4–11.9 years) for survivors, PFS and OS at 3 years were 60% (95% CI: 53–68%) and 73% (95% CI: 67–80%), respectively. Transplant-related mortality (TRM) was 4% both at 100 days and at 1 year post HDT-ASCT. Age and HCT-CI score were not associated with OS or PFS, and high HCT-CI did not correlate with TRM. Seven patients (4%) developed secondary myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML) at a median of 35 months (range 6–48) post HDT-ASCT. In this single-center cohort of elderly patients with NHL undergoing HDT-ASCT, this intervention is proved tolerable and effective, with results similar to historic controls in younger patients. Our data suggest that age alone should not preclude HDT-ASCT in elderly patients.

Introduction

High-dose therapy and autologous stem cell transplantation (HDT-ASCT) has been widely used as consolidative therapy in upfront or salvage settings across a range of non-Hodgkin lymphoma (NHL) histologies. The multicenter prospective randomized phase III PARMA trial established HDT-ASCT as a standard-of-care in relapsed or refractory intermediate or high-grade NHL chemosensitive to salvage chemotherapy, demonstrating both overall survival (OS) and progression-free survival (PFS) benefit¹. Similarly, the multicenter, randomized phase III CUP trial established PFS and OS benefit for early relapsed or refractory follicular lymphoma (FL) patients randomly assigned to HDT-ASCT compared to further chemotherapy alone². In mantle cell lymphoma (MCL), the prospective randomized trial of the European MCL Network, as well as the Nordic Lymphoma Group MCL2 study, established HDT-ASCT as part of first line therapy demonstrating long term PFS with HDT-ASCT in first remission^3–5. Prognosis is generally poor in T-cell lymphoma (TCL), due to low response rates and less durable remissions to conventional chemotherapy. Several retrospective and prospective studies have reported encouraging results with HDT-ASCT as first line consolidation option^6–9. The National Comprehensive Cancer Network (NCCN) recommends consideration of HDT-ASCT for most TCL in first complete remission¹⁰.

While the majority of patients diagnosed with NHL are greater than 60 years in age, most studies of HDT-ASCT for lymphoma have been performed on patients under the age of 60 years. Therefore, safety and efficacy of HDT-ASCT is not well established in this age group. Given the limited information on feasibility of HDT-ASCT in elderly^11–18, we retrospectively evaluated the outcomes of HDT-ASCT in 202 NHL patients aged 60 years and older at our institution over the 11-year period from January 2001 to December 2012. The aim of this study is to determine tolerability and efficacy of HDT-ASCT in elderly patients, with varied co-morbidities, toward the goal of clarifying eligibility for this procedure.

Patients and Methods

Patient and Graft Characteristics

This retrospective study was performed in patients 60 years of age or older with NHL who underwent HDT-ASCT at Memorial Sloan Kettering Cancer Center (MSKCC) between January 2001 and December 2012. The institutional Review Board at MSKCC approved data collection. Data was primarily extracted from comprehensive chart review and institutional databases. Pre-HDT-ASCT studies included physical examination, complete blood count, metabolic panel, chest radiography, computed tomography (CT) of the chest, abdomen and pelvis, fluorine-18-deoxyglucose positron emission tomography, bone marrow aspiration and biopsy, echocardiography, and pulmonary function testing (PFT). The hematopoietic stem cell transplant comorbidity index (HCT-CI)¹⁹ was calculated retrospectively through chart review of data immediately pre-HDT-ASCT. HCT-CI is defined as low for (0) points, intermediate for (1–2) points, and high for (>2) points. Peripheral blood hematopoietic stem cells (HPCs) were collected by leukapheresis following mobilizing cytokines with or without preceding chemotherapy priming^20–23. The minimal acceptable progenitor cell dose was 2×10⁶ CD34+ cells/kg. The day of HPC infusion was considered day 0. Patients received either filgrastim 5 mcg/kg/day from day +5 until neutrophil recovery (before 2006) or pegfilgrastim 6mg on day +1 (since 2006)²⁴. All patients were hospitalized for HDT-ASCT until engraftment unsupported by transfusions and adequate gastrointestinal recovery. Engraftment was defined as sustained neutrophil recovery that is the first of 3 consecutive days with a count > 0.5 × 10⁹/L. Patients were kept in single rooms and were managed clinically according to MSKCC standard guidelines, including infection prophylaxis.

Conditioning Regimens

HDT varied according to patient’s diagnosis, remission status, and co-morbidities. One hundred and forty-nine patients (74%) received BEAM (carmustine 300mg/m² × 1 dose, etoposide 100–150mg/m² every 12 hours × 8 doses, cytarabine 200mg/m² every 12 hours × 8 doses and melphalan 140mg/m² × 1 dose) as their primary HDT regimen. Several patients also received preceding immunotherapy with accelerated rituximab 375mg/m² × 2 doses (n=10; 5%) or radio-immunotherapy with Yttrium-90 ibritumomab tiuxetan (n=7; 3.5%) or Iodine-131 tositumomab (n=5; 2.5%) in addition to BEAM. Patients with central nervous system lymphoma (CNSL, n=18; 9%) received thiotepa 250mg/m² daily × 3 doses, busulfan 3.2mg/kg daily × 3 doses and cyclophosphamide 60mg/kg daily × 2 doses. Other less commonly used conditioning regimens included: melphalan and etoposide (n=6; 3%); cyclophosphamide and etoposide ± TBI (n=3; 1.5%); cyclophosphamide, carmustine and etoposide (n=2; 1%), mitoxantrone and melphalan (n=1; 0.5%); rituximab, carmustine, cyclophosphamide and etoposide (n=1; 0.5%). HDT was not routinely adjusted for body weight or impaired renal function. For those patients who received involved field radiation therapy (IFRT) prior to HDT-ASCT, radiation was limited to no more than two anatomically involved lymph node regions, was administered before HDT for disease measuring at least 5 cm before salvage therapy. IFRT was delivered in 1.5 Gy fractions twice daily to a total dose of 30 Gy over 10 days²⁵.

Statistical Methods

Descriptive statistics were used to summarize patient characteristics. Kaplan-Meier methodology and the logrank test were used to examine OS and PFS by patient characteristics. Cumulative incidence functions were used to estimate the cumulative incidence of progression of disease and TRM, with non-relapse mortality and progression as competing events, respectively. Analyses were performed in R statistical software version 2.13.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Patient demographics are summarized in Table 1. Sixty-seven percent were men. Of the 202 patients 60 years and above (median age 65, range 60–74 years), ninety patients were ages 60 – 64, 82 were ages 65 – 69 and 30 patients were 70 or older. Histologies for which patients underwent HDT-ASCT included diffuse large B-cell lymphoma (DLBCL, n=74, 37%), MCL (n=69, 34%), FL (n=12, 6%), CNSL (n=13, 7%), TCL (n=25, 12%). Nine patients (4%) had other NHL histologies including: chronic lymphocytic leukemia/small lymphocytic lymphoma (n=2), marginal zone lymphoma (n=2), plasmablastic lymphoma (n=1), lymphoplasmacytic lymphoma (n=2), T cell rich B cell lymphoma (n=1) and blastic plasmacytoid dendritic cell lymphoma (n=1). HDT-ASCT in first remission was performed in 93 patients (46%), majority of which for MCL (n=65, 71%). Of the 202 patients, 170 (84%) were in complete remission (CR) at the time of HDT-ASCT. With regards to treatment prior to HDT-ASCT, 45% of patients received 1 line of therapy, 48% received 2 lines, and 7% received more than 2 lines of therapy. One hundred and eighty three (91%) patients received anthracycline containing combination chemotherapy at some point during their treatment course. Of the relapsed or refractory DLBCL cases (n = 65), 45 (70%) received ICE-based (ifosfamide, carboplatin, etoposide) and 10 (15%) received DHAP-based (dexamethasone, cytarabine, cisplatin) salvage therapy, while 10 (15%) received other salvage programs. Of the first remission MCL transplant patients (n = 65), 47 (72%) received ICE-based sequential therapy following R-CHOP²⁶ (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) while 15 (23%) received R-CHOP in combination with cytarabine containing programs in either sequence or alternating cycles. The majority of these cases received induction with either high-dose cytarabine³ or DHAP⁵. Three patients (5%) received other regimens. Of the TCL first remission patients (n = 20), 10 (50%) received CHOP +/− etoposide and 10 (50%) received sequential CHOP/ICE.

Table 1.

Patient Characteristics Overall (n = 202).

Median (range) age at transplant	65 (60 – 74)
Gender (male)	136 (67%)
Histology
DLBCL	74 (37%)
CR1: 3
CR2: 49
CR3 or more: 4
PR: 18
MCL	69 (34%)
CR1: 61
CR2: 4
PR: 4
FL	12 (6%)
CNSL	13 (7%)
TCL	25 (12%)
Other NHL	9 (4%)
Conditioning
BEAM	149 (74%)
Rituximab-BEAM	10 (5%)
RIT - BEAM	12 (6%)
Other	31 (15%)
Transplants in 1st remission	93 (46%)
HCT-CI
Low (0)	42 (21%)
Intermediate (1–2)	45 (22%)
High (>2)	115 (57%)

Diffuse large B-cell lymphoma (DLBCL), Mantle cell lymphoma (MCL), Follicular lymphoma (FL), Central nervous system lymphoma (CNSL), T-cell lymphoma (TCL), Non-Hodgkin lymphoma (NHL), complete remission (CR), partial remission (PR), radio-immunotherapy (RIT), hematopoietic stem cell transplant comorbidity index (HCT-CI), BEAM (carmustine, etoposide, cytarabine, melphalan).

The median number of HPCs infused was 5.0 × 10ˆ6 CD34+ cells/kg (range 1.5–14.7). The median time to neutrophil engraftment was 10 days (range 7–16). One patient died before neutrophil engraftment. Forty-four patients (22%) were readmitted to the hospital in the first 100 days post HDT-ASCT (range 16–98 days). Majority (n=30, 68%) were admitted for infection, 9 (21%) for transplant-related toxicity and 5 (11%) for other causes. With a median follow-up of 3.6 years (range 0.4–11.9) for survivors, PFS and OS at 3 years were 60% (95% CI: 53–68%) and 73% (95% CI: 67–80%), respectively (Figure 1). No difference was found in OS and PFS between patients older than 65 and those between 60–64 years of age. Similarly, we found no difference in OS and PFS in patients age 70 or above when compared to those between ages 60 and 69 (Figure 2). At 3 years, PFS in patients receiving upfront transplant was 67% compared to 54% in those who received transplant for relapsed or refractory disease (p=0.03). OS in upfront transplant was 81% compared to 66% in relapsed or refractory setting (p=0.01). Patients with MCL had better outcomes compared to other NHL histologies, with 3 years OS of 89% versus 65% (p=0.002) and PFS of 76% versus 52% (p=0.001) respectively. HCT-CI score was not associated with OS or PFS (Figure 3).

Figure 1.

PFS and OS in all patients

Figure 2.

OS and PFS

Figure 3.

OS and PFS according to HCT-CI

Overall, 77 patients (38%) progressed [DLBCL (n=34, 46%), MCL (n=22, 32%), FL (n=4, 33%), PCNSL (n=1, 8%), TCL (n=12, 48%) and others (n=4, 44%)]. Median time to treatment failure was 13.6 months (range 1.4–106). Incidence of disease progression at one year post-HDT-ASCT was 28% (21 of 74) in DLBCL, 6% (4 of 69) in MCL, 17% (2 of 12) in FL, 8% (1 of 13) in PCNSL, 32% (8 of 25) in TCL and 22% (2 of 9) for all other subtypes.

Eight patients experienced TRM within the first 100 days post transplant. Five died from infections and 3 from treatment-related organ toxicity. Cumulative incidence of TRM was 4% (95% confidence interval [CI]: 2–7%) at 100 days. This remained the same at 1 year post-HDT-ASCT. High HCT-CI was not associated with TRM (Figure 4).

Figure 4.

TRM according to HCT-CI

Seventy-six (38%) patients died post HDT-ASCT. The most common cause of death was progression of disease, which accounted for 54 deaths (71%). Other deaths were attributable to: infections (n=5), organ toxicity (n=4), myelodysplastic syndrome (MDS) or acute leukemia (AML) (n=4) and solid tumors (n=4). Four patients died of unknown causes. One patient died from graft-versus-host disease following a subsequent allogeneic stem cell transplant.

Overall seven patients (4%) developed secondary MDS or AML, at a median of 35 months post-HDT-ASCT (range 6–48); other secondary malignancies (solid tumors) were seen in 9 patients (4.5%) following HDT-ASCT.

Demographics of patients age 70 or older at time of HDT-ASCT are summarized in Table 2. In this group, comorbidities that compose the HCT-CI were present in 73% of patients, the most common ones being prior malignancy and pulmonary or cardiac disease. Of the total thirty patients in this age group, 18 are alive, including 4 that have experienced disease progression. Causes of death in the other 12 patients were: progression of disease (n=7), secondary MDS/AML (n=3), secondary solid tumor (n=1) and infection (n=1). The incidence of MDS/AML in patients age 70 or older was 10% (n=3 out of 30 total), whereas its incidence in the ages 60–69 was 2% (n=4 out of 172 total) (p=0.07). As stated before, PFS, OS and TRM in this age group were not significantly different from patient between 60–69 years of age.

Table 2.

Patient Characteristics Age 70 or above: (n = 30).

Gender (male)	25 (83%)
Histology
DLBCL	12 (40%)
CR2: 8
CR3: 1
PR: 3
MCL	8 (27%)
CR1: 8
FL	4 (13%)
CR2: 3
CR3: 1
TCL	3 (10%)
CR1: 2
PR: 1
Other NHL: 2	3 (10%)
HCT-CI
Low (0)	8 (27%)
Intermediate (1–2)	3 (10%)
High (>2)	19 (63%)

Diffuse large B-cell lymphoma (DLBCL), Mantle cell lymphoma (MCL), Follicular lymphoma (FL), Primary central nervous system lymphoma (PCNSL), T-cell lymphoma (TCL), Non-Hodgkin lymphoma (NHL), complete remission (CR), partial remission (PR), hematopoietic stem cell transplant comorbidity index (HCT-CI)

Discussion

HDT-ASCT for NHL has not been commonly offered to elderly patients perhaps due in part to limited data on feasibility and efficacy of this treatment modality in older individuals. Given the demonstrable prognostic impact of HDT-ASCT in younger patients, as previously highlighted by prospective randomized studies, the omission of HDT-ASCT based upon age alone could potentially affect survival in this group. We demonstrate in this study, that HDT-ASCT is safe and effective in selected patients from age 60 to 74 years, and that age alone is not a significant risk factor for TRM, even in the subgroup age 70 to 74. We observed a 3 year OS and PFS of 73% and 60% respectively. Our outcomes appear similar or favorable, compared to other published reports on elderly patients with NHL undergoing HDT-ASCT^16–18. This is most notable when compared to study by Chihara et al¹⁵ (n = 484) that showed a 2-year PFS and OS of 48% and 58% respectively. Other studies are summarized in Table 3. These varying results in the survival rate are likely attributable to differences in patient and disease characteristics, as we included patients who received HDT-ASCT as consolidation therapy in first remission for a relatively indolent disease (MCL), in contrast to relapsed and refractory aggressive NHL. Approximately one-third of our patients had MCL, with the majority undergoing HDT-ASCT in first remission. This group had a superior OS and PFS at 3 years compared to the other NHL histologies (OS: 89% versus 65% and PFS: 76% versus 52%). Again, this is likely attributed both to the natural history of MCL and to the fact that patients receiving consolidation in first remission are typically less heavily pretreated. In our study 93 (46%) patients received HDT-ASCT as consolidative treatment in first remission. The majority of patients had received 2 or less lines of therapy, and were in remission prior to HTD-ASCT. Certainly, chemosensitivity prior to HDT-ASCT carries the largest prognostic relevance^{27, 28}.

Table 3.

Comparative Table of Outcomes of HDT-ASCT in Elderly with NHL

Reference	Patients	Prep regimen	Early TRM 100 days	Late TRM 1 year	OS (%)	PFS (%)	HCT - CI	MDS AML
Jantuen et al Haematolog 2008	N=463 DLBCL Age ≥ 60	BEAM 73%	4.4%	8.7%	3 year 60%	3 year 51%	Not reported	3 year: 3.5%. 5 year: 7.5%.
Jantuen et al Ann Onco 2012	N=79 MCL Age ≥ 65	BEAM 72%	3.8%	3.8%	5 year 61%	5 year 29%	Not reported	Not reported
Hosing et al Ann Onco 2008	N=99 NHL Age ≥ 65	BEAM 90%	3%	8% 26 mo	3 year 61%	3 year 48–69	No effect	8 patients median 20 months
Chihara et al BBMT 2014	N=484 DLBCL Age ≥ 60	Variable	4.1%	5.9%	2 year 58%	2 year 48%	Not reported	Not reported
MSKCC Data	N=202 NHL Age ≥ 60	BEAM 74%	4%	4%	3 year 73%	3 year 60%	No effect	7 (4%) patients median 35 months

Diffuse large B-cell lymphoma (DLBCL), Mantle cell lymphoma (MCL), Non-Hodgkin lymphoma (NHL), BEAM (carmustine, etoposide, cytarabine, melphalan), Transplant-related mortality (TRM), Overall survival (OS), Progression-free survival (PFS), Hematopoietic stem cell transplant comorbidity index (HCT-CI), Myelodysplastic syndrome (MDS), Acute myeloid leukemia (AML)

Early TRM rate at day 100 was 4% in our study, which is comparable to those reported in other earlier series^{13, 16–18}. No additional TRM events were noted at 1 year post HDT-ASCT, likely reflecting the immediate and relatively short-term risk of barrier mucocutaneous toxicity in the neutropenic state post-HDT-ASCT. This low rate of TRM likely reflects both patient selection prior to HDT-ASCT and ongoing advances in supportive care. Contextually, this data further highlights the overwhelming majority of treatment failure events being progression of primary NHL.

Interestingly, the incidence of MDS/AML in patients age 70 or older was higher than those between ages 60–69 (p=0.07). This could likely be attributed to higher incidence of MDS/AML with increasing age. Previous studies reported older age, radiation and the quantity of pre-transplant therapy as risk factors for therapy-related MDS and AML post HDT-ASCT^29–31, however in those studies older age was defined as age older than 35–40 years. Our study suggests that patients older than 69 may be at a higher risk for secondary MDS/AML and this may indicate the need for evaluation of bone marrow for early MDS prior to stem cell collection.

Within this cohort, we analyzed other clinical characteristics that may predict adverse outcomes after transplant. HCT-CI, which derives from physiologic assessment and the presence of particular co-morbidities at the time of transplant, is a useful tool to predict survival after allogeneic transplant¹⁹. Several retrospective analyses assessed the role of HCT-CI as a predictor of outcomes in patients undergoing HDT-ASCT^{17, 32, 33}. In our study, although most of the patients had high HCT-CI score, this score was not associated with risk of TRM, nor was it associated with either OS or PFS. Our results are consistent with the study by Hosing et al.¹⁷ (n = 99) reporting no effect of HCT-CI on HDT-ASCT outcomes in patients older than 65 with relapsed and refractory NHL. However, we note that the lack of demonstrable prognostic impact of HCT-CI on HDT-ASCT outcomes may be due to lack of statistical power with limited patient numbers, as compared to the CIBMTR study³³. Other tools, such as comprehensive geriatric assessment, could potentially be utilized to identify elderly-frail patients at higher risk of post-HDT-ASCT mortality³⁴. We did not incorporate a geriatric assessment in patient selection.

Our study has a number of limitations. It is retrospective and conducted in a single center. Small number of patients included in this study overall, and specifically in the group older than 69, could have contributed to the lack of demonstrable prognostic impact of age and HCT-CI on HDT-ASCT outcomes in this population. Another important limitation pertains to analyzing transplant outcomes in a rather non-homogeneous cohort of patients, ranging from those with more indolent histologies who received HDT-ASCT as consolidation therapy in first remission, to other more aggressive NHL subtypes who received HDT-ASCT following multiple lines of therapy for relapsed and refractory disease.

In conclusion, within this cohort of elderly patients with NHL undergoing HDT-ASCT, we demonstrate feasibility and efficacy with survival comparable to younger historical controls^{1, 2, 5, 35}. Our data suggests that age alone is not predictive of HDT-ASCT outcomes and therefore should not be used to preclude HDT-ASCT up to age 74. While further prospective investigation is warranted, particularly in patients older than 74 years of age, our data do not support routine exclusion of patients based on age or HCT-CI, especially in clinical settings where clear benefit has been defined and the vast majority of survival events are secondary to disease progression^{1–6, 9, 33}. Although we demonstrate relatively favorable results post HDT-ASCT in elderly patients, application of this treatment should be carefully considered individually and reserved for a selected population. Further prospective investigation toward identifying a subset of elderly patients who carry a high risk of treatment-related morbidity or mortality or of MDS/AML is warranted with careful consideration toward NHL disease risk and potential benefit of HDT-ASCT.

Highlights.

• High dose therapy- autologous stem cell transplant (HDT-ASCT): feasible in elderly.

• HDT-ASCT outcomes for NHL in elderly are comparable to younger historic controls.

• Age alone, should not be used to preclude HDT-ASCT in elderly up to age 74.

• HCT-CI score was not associated with transplant outcomes in elderly.