BEAM versus Pharmacokinetics-Directed BuCyVP16 Conditioning for Patients with Peripheral T-cell Lymphoma Undergoing High Dose Therapy with Autologous Hematopoietic Cell Transplantation

Graphical Abstract

To the Editor:

Despite differences in underlying pathobiology, the nodal T-cell lymphomas—referring to peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS), nodal T-follicular helper (TFH) cell lymphomas, and anaplastic lymphoma kinase-negative anaplastic large cell lymphoma (ALK– ALCL)—are treated similarly in the upfront setting with anthracycline-based combination chemotherapy. While response rates to induction are high, durability of responses are suboptimal and relapses are common. ¹ These results have prompted efforts at consolidation with high dose therapy and autologous hematopoietic cell transplantation (HDT/AHCT) in first remission. While randomized trials comparing HDT/AHCT versus observation have not resulted, non-comparative phase II trials and observational reports have examined the impact of HDT/AHCT, generally demonstrating improved progression-free (PFS) and overall survival (OS) compared to historic, non-consolidative approaches.²

Borrowing from data on its use in non-Hodgkin B-cell lymphomas, BEAM is the most common conditioning regimen used in PTCL. However, other conditioning regimens have been explored, including gemcitabine- and busulfan-based combinations. In particular, BuCyVP16 (busulfan, cyclophosphamide, etoposide) is a frequently used conditioning regimen, traditionally administered via weight-based dosing (WBD), though later studies using pharmacokinetic (PK)-directed dosing to guide exposure showed improvements in relapse rates and PFS in non-Hodgkin lymphoma.³ While comparison between BuCyVP16 and other regimens have been performed in non-Hodgkin lymphoma, few patients with T-cell lymphoma have been included in these reports. To compare the efficacy of BEAM versus PK-directed ByCyVP16 in T-cell lymphoma, we performed a bi-institutional study in patients with nodal PTCL who underwent HDT/AHCT in first complete remission (CR). Direct comparisons of alternative conditioning regimens to BEAM have not been performed in a dedication fashion for nodal T-cell lymphomas.

We performed a retrospective study of patients with nodal T-cell lymphoma who underwent HDT/AHCT in first CR using patients from the transplant databases at Memorial Sloan Kettering (MSK) and the Cleveland Clinic. We included patients ≥ 18 years of age at diagnosis who underwent HDT/AHCT on or after January 1, 2013. Histology was limited to PTCL-NOS, nodal TFH cell lymphomas (including angioimmunoblastic T-cell lymphoma), and systemic ALK– ALCL. Analysis was limited to patients who achieved a first CR after induction as per chart documentation and proceeded to upfront HDT/AHCT. Patients underwent HDT/AHCT as per standard institutional procedures. At MSK, BEAM is the standard conditioning approach, whereas at the Cleveland Clinic, PK-directed BuCyVP16 has been a standard approach, as previously described.³

Based on the conditioning regimen received, two cohorts of patients were analyzed: BEAM versus BuCyVP16. The primary endpoints were comparing PFS and OS (from day of transplant) between the two groups. PFS was defined as the time from transplant to progression, death, or last follow-up. OS was defined as the time from transplant to the time of death or last follow-up. Secondary endpoints included length of hospital stay, time to engraftment, and non-relapse mortality. Length of stay estimation was restricted to patients who received inpatient transplants only.

Baseline characteristics for the two cohorts are shown in Table 1. Seventy-five patients received BEAM and 24 patients received BuCyVP16. The cohorts were similar with respect to age, sex, histology, stage, and International Prognostic Index. The most common histologies in both cohorts were PTCL-NOS and angioimmunoblastic T-cell lymphoma. In the BEAM cohort, more patients received BV-CHP, likely reflecting prior participation in the ECHELON-2 trial at MSK. In addition, there were more patients in the BEAM cohort who received ‘other’ induction regimens, reflecting prior MSK participation in two clinical trials of alternative induction therapy in T-cell lymphomas (oral azacytidine plus CHOP [NCT03542266] and lenalidomide plus CHOEP [NCT02561273]).

Table 1.

Baseline Characteristics.

Characteristic	BEAM, N=751	BuCyVP16, N=241	p-value2
Age at diagnosis	61 (51, 67)	56 (49, 68)	0.7
Sex  Female  Male	32 (43%) 43 (57%)	7 (29%) 17 (71%)	0.2
Histology  AITL  ALK– ALCL  PTCL-NOS	43 (57%) 9 (12%) 23 (31%)	8 (33%) 6 (25%) 10 (42%)	0.086
Stage  I  II  III  IV	2 (2%) 6 (8%) 29 (39%) 38 (51%)	1 (4%) 2 (8%) 9 (38%) 12 (50%)	>0.9
IPI  0–1  2–3  4–5	7 (9%) 59 (79%) 9 (12%)	4 (17%) 16 (66%) 4 (17%)	0.4
Induction  CHOP  CHOEP/EPOCH  BV-CHP  Other	11 (15%) 38 (51%) 12 (16%) 14 (18%)	10 (42%) 11 (46%) 2 (8%) 1 (4%)	0.028

¹Median (IQR); N (%)

²Wilcoxon rank sum test; Pearson’s Chi-squared test; Fisher’s exact test.

AITL, angioimmunoblastic T-cell lymphoma; ALK– ALCL, anaplastic lymphoma kinase-negative anaplastic large cell lymphoma; BEAM, carmustine, etoposide, cytarabine, melphalan; BV-CHP, brentuximab vedotin, cyclophosphamide, vincristine, prednisone; BuCyVP16, busulfan, cyclophosphamide, etoposide; CHOP, cyclophosphamide, doxorubicin, vincristine, prednisone; CHOEP/EPOCH, cyclophosphamide, doxorubicin, vincristine, etoposide, prednisone; IPI, International Prognostic Index; PLT, platelet count; PTCL-NOS, peripheral T-cell lymphoma, not otherwise specified.

The median time to neutrophil engraftment was similar between the two cohorts (p=0.14), while the median time to platelet engraftment was longer for BEAM (21 days, IQR: 18, 23) versus BuCyVP16 (16 days, IQR: 15, 18; p<0.001). The median (IQR) length of stay in the BEAM cohort was 22 days (20, 25) versus 21 (21, 22) days in the BuCyVP16 cohort (p=0.2).

Survival comparisons are shown in Figure 1. PFS was similar between the two cohorts, with a median PFS of 4.4 years (95% CI: 1.6, not reached [NR]) in the BEAM cohort versus 4.8 years (2.0, NR) in the BuCyVP16 cohort (p=0.8). The five-year PFS estimates in the BEAM versus BuCyVP16 cohorts were 49% (95% CI: 38%, 63%) versus 41% (95% CI: 23%, 72%), respectively. With a median follow-up time of 4.6 and 5.6 years among survivors in the BEAM and BuCyVP16 cohorts, the median OS in the BEAM versus BuCyVP16 cohorts was 8.5 years (95% CI: 6.2, NR) and NR (95% CI: 2.7, NR), respectively (p=0.6). Five-year OS estimates in the BEAM versus BuCyVP16 cohorts were 67% (95% CI: 56%, 80%) versus 66% (95% CI: 49%, 88%), respectively. The 1- and 2-year non-relapse mortality rates were low in each group at 1.3% and 2.7% (BEAM) and 0.0% and 4.4% (BuCyVP16).

Figure 1. Progression-Free and Overall Survival.

Progression-free survival (Figure 1A) and overall survival (Figure 1B) for patients with nodal peripheral T-cell lymphoma conditioned with BEAM versus BuCyVP16.

There are limited data to guide the choice of conditioning regimens to use prior to AHCT in T-cell lymphomas. Therefore, we performed this retrospective comparison between BEAM and BuCyVP16, two of the most common conditioning regimens used in the management of lymphoma. Dedicated comparison of these regimens for T-cell lymphomas has not been performed.

BEAM is the most utilized conditioning regimen in all lymphomas, including T-cell lymphomas. BuCyVP16 as a conditioning regimen was developed over two decades ago given the known effectiveness of busulfan in various lymphoproliferative disorders—multiple subsequent reports demonstrate the efficacy of this program, primarily in B-cell non-Hodgkin lymphoma and Hodgkin lymphoma.⁴ PK-directed BuCyVP16 uses real-time blood samples to individualize busulfan dosing in attempts to maximize the proportion of patients receiving optimal drug exposure. While comparisons between BEAM and WBD or PK-directed BuCyVP16 have been performed for B-cell lymphomas, only minimal data exists for T-cell lymphomas. The results presented herein found no statistically significant differences in PFS or OS among those conditioned with BEAM versus PK-directed BuCyVP16. In addition, there were no statistically significant differences in length of study or neutrophil engraftment time, though there was a slight difference in platelet recovery time favoring BuCyVP16. This reasoning for this is unclear and this modest difference likely has no meaningful clinical impact.

Our results are generally consistent with prior, similar comparisons. A large-scale effort comparing patients with non-Hodgkin lymphoma, primarily diffuse large B-cell lymphoma, conditioned with BEAM versus WBD BuCyVP16 between 2006 and 2014 found no difference in PFS, OS, or cumulative incidence of relapse between the two regimens.⁵ On multivariable analysis accounting for age, gender, race, prior treatment remission status, transplant year, comorbidity index, length of stay, and length of time to diagnosis, no statistically significant differences in PFS or cumulative incidence of relapse were observed, though a significant improvement in OS favoring BEAM (HR 1.56; 95% CI, 1.16–2.10) was detected.

It is worth highlighting the survival observed in this report. Historical data from the late 20^th century in the International T-cell Lymphoma Project, a collaborative effort including over 1000 patients from twenty-two global institutions, reported five-year FFS rates of 20%, 18%, and 36% for PTCL-NOS, angioimmunoblastic T-cell lymphoma, and ALK– ALCL, respectively.¹ In addition, five-year OS rates for the same histologies were 32%, 32%, and 49%, respectively.¹ Noting the bias inherent in our report (limited to patients achieving CR and proceeding to HDT/AHCT) our results are promising, with five-year PFS and OS of 49% and 67% among the BEAM cohort. We suspect the observed improvements are secondary to the use of brentuximab vedotin in CD30-positive disease, and potentially an increased use of etoposide-containing induction regimens, use of HDT/AHCT in first remission, use of allogeneic transplantation in salvage settings, and the approval of multiple agents for relapsed and refractory disease. Our results, in conjunction with long-term follow-up of the Nordic Lymphoma Group-T-01 trial show that anthracycline-based induction followed by HDT/AHCT in first CR can result in durable remissions in a subset of patients with nodal-based T-cell lymphomas. The ongoing, randomized TRANSCRIPT trial will specifically test the role of HDT/AHCT in patients with nodal-based T-cell lymphoma in first CR and utilizes the BEAM conditioning regimen (NCT05444712).

Whether alternative conditioning regimens may improve upon BEAM in the management of T-cell lymphomas is unknown. A small study of CHOEP induction followed by gemcitabine/busulfan/melphalan HDT/AHCT in newly diagnosed PTCL closed early after five patients enrolled due to lack of efficacy (NCT01746173).⁶

Our study has limitations inherent to its retrospective nature. In addition, while this was a bi-institutional study, the relatively small number of patients in each cohort limits the ability to make histology-specific comparisons. In addition, detailed adverse event data, such as rates of oral and gastrointestinal mucositis, was unavailable for comparison and would be of interest. We found that time to platelet engraftment was significantly different between the cohorts, but the reason for this difference is unclear. Despite these limitations, these data align with previous literature on the safety and effectiveness of both regimens across lymphoma subtypes. Moreover, the efficacy and widespread use of BEAM conditioning for all lymphomas across the United States and internationally argues for harmonization in its use across centers as a standard of care to standardize future studies.

Data Availability:

Data presented herein is available upon reasonable request to the corresponding author.