Hematopoietic progenitor cell harvesting is feasible after treatment with Brentuximab vedotin in CD30+ lymphoma patients who received multiple prior lines of treatment

Abstract

Brentuximab vedotin (BV), an antibody-drug conjugate that targets CD30, induces high response rates in CD30+ lymphoid malignancies. It is unknown if BV use affects procurement of autologous CD34+ stem cells and hematopoietic engraftment after ASCT. We examined 42 patients treated with BV before mobilization. Median times from diagnosis to transplant, from initial BV treatment to transplant and from last BV treatment to stem cell collection were: 21 months (range, 10-210), 5 months (range, 1.5-16.8), and 30 days (range, 2-280), respectively. Mobilization was successful on first attempt in 38 patients (90.4%). The median number of infused CD34+ cells was 5.46×10⁶/kg (range, 1.65-54.78×10⁶/kg). The median times to neutrophil and platelet engraftment were 10 (range, 9-13), and 10.5 days (range, 7-35), respectively. BV before HDC-ASCT did not adversely affect peripheral blood stem cell mobilization and subsequent engraftment in a cohort of heavily pre-treated patients with CD30+ lymphomas.

Introduction

High dose chemotherapy followed by autologous stem cell transplant (ASCT) is an established standard for treatment of relapsed Hodgkin and non-Hodgkin lymphoma ¹. Successful engraftment depends largely on the collection of an adequate number of hematopoietic progenitor cells (HSCs). While the optimal dose for rapid neutrophil and platelet engraftment is considered to be >5×10⁶ CD34+ cells/kg ^2,3, 5 to 40% of patients fail to collect a minimum of 2×10⁶ CD34+ cells per kilogram ^4-6. Specific antineoplastic agents have been associated with poor mobilization rates, including lenalidomide ^7,8, fludarabine ^9,10 and melphalan¹¹. Recent studies have shown that stem cell mobilization is impaired in patients with diabetes or glucose intolerance ^12,13.

The antibody – drug conjugate Brentuximab vedotin (BV) has been shown to result in high response in CD30+ malignancies ^14,15. Phase II studies showed BV results in overall response rates of 75% in relapsed/refractory Hodgkin Lymphoma ¹⁶ and 86% in relapsed refractory anaplastic large cell lymphoma (ALCL) ¹⁷.

When used as salvage before reduced-intensity allo-HSCT, BV did not appear to adversely affect engraftment, incidence of GVHD or survival ¹⁸. In a follow-up report, BV was associated with improved progression-free survival and reduced transplant related mortality ¹⁹.

Initial studies have explored the use of BV in combination with chemotherapy for first line treatment of HL ²⁰ and CD30+ peripheral T cell lymphomas ²¹. Ongoing studies are investigating the use of BV alone or in combination with chemotherapy for salvage of CD30+ lymphomas in first relapse.

It is unknown whether use of BV before hematopoietic cell mobilization would affect collection of CD34+ stem cells and subsequent engraftment. We therefore examined the outcomes 42 patients who were treated with BV prior to HDC-ASCT.

Methods

We retrospectively reviewed the HDC-ASCT databases of University Hospitals Case Medical Center (UHCMC) and MD Anderson Cancer Center (MDACC). Forty-two patients who were treated with BV prior to HDC-ASCT between February 2009 and April 2014 were included. BV treatment was prescribed as part of standard of care for HL and ALCL or within clinical studies including DLBCL and other lymphomas. Clinical data collected included age, gender, histologic diagnosis, complete blood count, red blood cell (RBC) and platelet transfusion history, time from diagnosis to transplant, time from initial BV treatment to transplant, time from last BV treatment to stem cell collection, history of prior therapies and number of mobilization procedures. Chemotherapy/G-CSF in first mobilization was standard at MDACC, whereas Plerixafor/G-CSF was used as first mobilization regimen at UHCMC. Successful peripheral blood hematopoietic progenitor cell (HPC) mobilization was defined as achieving a target of 2×10⁶ CD34+ cells/kg. A cohort of 125 lymphoma patients that underwent ASCT without prior BV treatment between 2004 and 2014 at UHCMC was used as a control for mobilization yield.

Differences between groups were examined with the student t test. Correlations were studied with the Pearson product-moment correlation coefficient. Response was assessed using standard criteria ²². Overall survival and actuarial event-free probabilities were calculated by the Kaplan-Meier method ²³.

Results

Median age was 37 years (range, 18-67); 52% (n=22) were male (Table 1). Diagnoses were HL (n=30; 71) and NHL (n=12; 29%; anaplastic large cell, n=6; diffuse large B-cell, n=3; unknown subtype, n=3). Median times from diagnosis to transplant, from initial BV treatment to transplant and from last BV treatment to stem cell collection were: 21 months (range, 10-210), 5 months (range, 1.5-16.8), and 30 days (range, 2-280), respectively. Our subjects had failed multiple conventional treatments with a median of 3 (range, 2–8) lines of treatment before HDC-ASCT. Fifteen patients (35%) received prior radiation therapy; radiation fields included mediastinum (n=7), mantle (n=3), neck (n=2), lower extremity (n=1), and right acetabulum (n=1). The site of radiation was not available for 2 patients. One subject had radiation to both mantle field and right acetabulum. BV was given at 1.8 mg/kg IV every 21 days. Median number of BV cycles was 4 (range, 1-16).

Table 1.

Baseline characteristics of the patient population

Median age, years (range)	37 (18-67)
Diagnosis, n (%)
Hodgkin lymphoma	30 (71)
Non Hodgkin Lymphoma	12 (29)
Anaplastic large cell lymphoma	6 (14)
Diffuse large B-cell lymphoma	3 (7)
Other	3 (7)
Median number of prior lines of therapy	3
Previous radiation therapy, n (%)	15 (35)
Best response to brentuximab, n (%)
CR	23 (55)
PR	7 (17%)
Median time from diagnosis to transplant, months (range)	21 (10-210)
Median time from last BV treatment to transplant, months (range)	5 (1-11)
Median time from last BV treatment to HPC collection, days (range)	30 (2-280)

BV: brentuximab vedotin; CR, complete response; HPC: hematopoietic progenitor cells;

PR, partial response;

The overall response rate to BV treatment was 72% (CR 55% + PR 17%). Thirty patients (71%) were in complete remission (CR) at the time of transplant (CR2 = 6; CR≥3 = 24), 4 (10%) were in partial remission (PR) (PR2 = 1; PR≥3 = 3), 6 patients (14%) had stable disease and 2 patients (5%) were transplanted with progressive disease.

The mobilization regimens included chemotherapy/G-CSF in 32 patients (76%) and Plerixafor/G-CSF in 10 patients (24%). Thirty-eight (90.4%) of 42 patients were successfully mobilized on the first attempt. Three subjects (7.1%) required a second mobilization regimen for successful collection. Second mobilization regimens included cyclophosphamide/G-CSF (n=2) and Plerixafor/G-CSF (n=1). One patient proceeded to ASCT after first mobilization yielded a borderline graft content of 1.65×10⁶ CD34+ cells/kg.

The median number of infused CD34+ cells for the whole cohort was 5.46 ×10⁶/kg (range, 1.65-54.78 ×10⁶/kg). Because the first line mobilization regimen was different between centers, we compared the number of CD34+ cells collected according to mobilization agent. There was no statistically significant difference in the CD34+ cell dose of patients mobilized with first line Plerixafor/GCSF (median, 4.81×10⁶/kg), or with chemotherapy and G-CSF-based regimens (median, 5.53×10⁶/kg)(p=0.35).

There was no statistically significant correlation between the number of cycles of BV administered and number of CD34+ cells/kg (p=0.59). The median number of CD34+ cells collected was 5.5×10⁶/kg and 5.24×10⁶/kg for patients receiving 5 or more cycles, versus 4 cycles or less (p=0.45).

We also compared collection results after BV with those of a cohort of patients that underwent autologous transplant for relapsed lymphoma who did not receive BV (n=125). There was no difference in the median number of cells collected with prior BV (mean = 6.9×10⁶/kg; median = 5.46×10⁶/kg) and without previous BV (mean = 5.78×10⁶/kg; median = 5.1×10⁶/kg), (p=0.38). Mobilization yields remained comparable when we excluded patients that had HPCs collected after Plerixafor – containing mobilization regimens (p=0.58). Similarly, prior radiation therapy did not result in lower HPC collection (median CD34+ cell dose: prior radiation = 5.45×10⁶/kg; no prior radiation = 5.72×10⁶/kg; p=0.46).

All patients engrafted neutrophils and platelets at a median time of 10 days (range, 9-13), and 10.5 days (range, 7-35), respectively. The median time to RBC transfusion independence was 8 days (0-34). With a median follow-up of 12 months, day 100 and one-year treatment related mortality rates were 0%. Although patients in our cohort were heavily pre-treated, one-year actuarial event-free and overall survival was 52.7% and 81.6%, respectively (Figure 1).

Figure 1.

Progression free (solid line) and overall survival (dashed line) after ASCT of 42 patients treated with Brentuximab vedotin before autologous hematopoietic cell mobilization.

Discussion

A possible effect of BV in the mobilization of HPCs prior to ASCT has not been examined before. In this report, we show HPC harvesting is feasible after treatment with BV in patients that received multiple prior lines of chemotherapy. More than 90% of our patients were successfully mobilized on the first attempt. Accordingly, hematopoietic engraftment was not affected by prior BV therapy. Neutrophil and platelet engraftment times were comparable to those reported after ASCT in lymphoid malignancies ²⁴.

This report has limitations, in particular its retrospective nature and the inclusion of a heterogeneous patient population. In addition, the majority of patients received a small number of cycles of BV prior to mobilization (median of 4 cycles); it is possible that subjects with more prolonged exposure to this agent may have impaired stem cell mobilization. However, in our patient cohort there was no correlation between the number of BV cycles and HPC mobilization yields, and CD34+ cell doses were comparable to those obtained from patients without exposure to BV. Furthermore, if BV is used in salvage before ASCT, it is likely that the goal will be to administer only enough cycles to achieve remission, a strategy that is represented in our patient population. While radiation did not affect HPC mobilization here, none of our subjects received extensive radiation to hematopoietic sites. Several trials are investigating the use of BV combined with chemotherapy for the treatment of lymphoid malignancies, and the effect of these combinations on stem cell mobilization will need to be evaluated in future studies.

The early post – transplant disease control is likely reflective of the high activity of BV against CD30+ lymphoid malignancies. The only patient who relapsed within 100 days of ASCT had a bulky mediastinal mass and was later treated with radiation and allogeneic HSCT.

With the advent of new, targeted therapies, it remains important to demonstrate that they do not represent a hindrance to subsequent therapeutic plans. The high rates of response and excellent tolerability of BV will likely lead to its use in earlier lines of therapy. Our results would suggest BV does not adversely affect HPC mobilization and subsequent engraftment.

• HPC mobilization is feasible after brentuximab vedotin.

• Brentuximab vedotin did not result in a decrease in HPC yield after mobilization.

• Engraftment was not impaired after ASCT after treatment with brentuximab vedotin.