Most patients undergoing high-dose therapy and autologous transplant for multiple myeloma eventually develop a disease recurrence. However, to the authors' knowledge, the optimal salvage treatment for these patients is not well defined. Both autologous and allogeneic hematopoietic stem cell transplantations have been used for salvage therapy. The outcomes of salvage autologous or allogeneic transplants were analyzed retrospectively in patients relapsing after an autograft.
Fourteen patients (median age, 52 yrs) received a second autograft for salvage, whereas 26 patients (median age, 51 yrs) underwent a reduced-intensity allogeneic transplantation (related in 18 patients and unrelated in 8 patients). The median interval between the first and the second transplant was 25 months in the autologous group and 17 months in the allogeneic group. The two groups were evenly matched with regard to other disease characteristics.
After a median follow-up of 18 months for the autologous group and 30 months for the allogeneic group, the median progression-free survival (PFS) and overall survival (OS) in the 2 groups were 6.8 months versus 7.3 months and 29 months versus 13 months, respectively. Acute and chronic graft versus host disease (15%) was the most common cause of nonrecurrence mortality in the allogeneic group and infections (14%) in the autologous group. On univariate analysis, an interval of > 1 year between the first and the salvage transplant predicted a better OS in the allogeneic group.
High-dose chemotherapy (HDT) with autologous stem cell transplantation (ASCT) results in higher complete remission rates, overall survival (OS), and event-free survival (EFS) in patients with multiple myeloma (MM).1–3 Two large randomized trials have shown the survival advantage of HDT over conventional dose therapy (CDT), with a median increase in EFS and OS by approximately 12 months in the HDT arm. However, this approach is not curative and most patients undergoing autografting for MM eventually develop disease recurence.4
To our knowledge, here is no standard salvage treatment for patients who develop disease recurrence after an autograft. Treatments ranging from conventional agents to salvage autologous and allogeneic transplants have been used.5–11
Mehta et al.10 reported the results of salvage autografts that were associated with a complete remission rate of 41%, a 3-year probability of EFS of 20%, a 3-year probability of OS of 54%, and a 1-year transplant-related mortality (TRM) of 10.5%. They found that OS was superior with autografts, but disease progression was significantly lower with allogeneic transplantation.
Allogeneic transplantation may lower the risk of recurrence due to infusion of myeloma-free graft and the potential for a graft-versus-myeloma effect mediated by donor T-lymphocytes.12 This effect was evident in the disease remissions induced by donor lymphocyte infusions in patients.13 Allogeneic transplantation using myeloablative conditioning regimens is associated with 30–50% TRM in most published reports.10, 14, 15 The emergence of reduced intensity preparative regimens has renewed interest in allogeneic transplant strategies in myeloma due to the lower regimen-related toxicity without a compromise in the graft-versus-myeloma effect.16–18
In the current study, we retrospectively analyzed the outcomes after salvage autologous or reduced-intensity allogeneic transplants in patients with disease recurrence after an autograft.
MATERIALS AND METHODS
Fourteen patients received a second autograft for salvage between January 1992 and May 2004, whereas 26 patients underwent reduced-intensity allogeneic transplantation (related in 18 patients and unrelated in 8 patients) between June 1995 and May 2004. In general, younger patients (up to age 65 yrs) with available human leukocyte antigen-matched donors, financial clearance, better performance status, and less comorbidity were treated with an allogeneic transplant. Other patients with recurring disease, age ≤ 70 years, who were not eligible for an allogeneic transplant but still fulfilled the eligibility criteria for HDT, received a salvage autograft. They were treated at the University of Texas M. D. Anderson Cancer Center (MDACC) on institutional review board-approved protocols and provided written informed consent before treatment. Diagnosis of MM was made according to the criteria of the Chronic Leukemia Myeloma Task Force.19 The response criteria used were those defined by the European Group of Blood and Marrow Transplantation (EBMT), the Autologous Bone Marrow Transplantation Registry (ABMTR), and the International Blood and Marrow Transplantation Registry (IBMTR(.20 Responses were assessed at Day 90–100 after transplantation.
Eligible patients were age ≤ 70 years, with adequate cardiac (left ventricular ejection fraction > 45%), pulmonary (diffusing capacity of the lung for carbon monoxide > 50%), and hepatic function (bilirubin, transaminases < 2 times upper limit of normal). Preparative regimens were melphalan alone at a dose of 100 mg/m2 intravenously (i.v.) × 2 days (in 6 patients); melphalan at a dose of 70 mg/m2 i.v. × 2 days plus total body irradiation at 850 centigrays (cGy) (in 2 patients); topotecan at a dose of 3 mg/m2 i.v. × 5 days plus melphalan at a dose of 70 mg/m2 i.v. × 2 days plus cyclophosphamide at a dose of 1 g/m2 i.v. × 3 days (in 2 patients); thiotepa at a dose of 150 mg/m2 i.v. × 3 days plus busulfan at a dose of 0.8 mg/kg i.v. × 10 doses and cyclophosphamide at a dose of 60 mg/kg i.v. × 2 days.21 The source of hematopoietic stem cells, collected and cryopreserved before the first autograft, was peripheral blood in 12 patients and bone marrow in 2 patients.
Patients in the allogeneic were age ≤ 65 years and, in addition to the organ system function described for the autologous group, they also had adequate renal function (serum creatinine of < 2 mg/dL). Related and unrelated donors were acceptable if they matched at a minimum 5 of 6 loci on serologic typing and 9 of 10 loci on molecular typing, with a single mismatch allowed for class A and DQ loci. All patients who underwent transplantation after 1995 underwent high-resolution, molecular human leukocyte antigen typing. Preparative regimens were: fludarabine at a dose of 30 mg/m2 × 4 days plus melphalan at a dose of 70 mg/m2 i.v. × 2 days (in 24 patients); and fludarabine at a dose of 25 mg/m2 i.v. × 5 days plus cyclophosphamide at a dose of 1 g/m2 i.v. × 3 days (in 2 patients). Patients undergoing unrelated donor transplants also received rabbit antithymocyte globulin at a dose of 3 mg/kg × 4 days. Patients received tacrolimus for up to 6 months after a related transplant and for up to 9 months after an unrelated transplant. Methotrexate at a dose of 5 mg/m2 was given on Days 1, 3, and 6 after a related transplant, and on Days 1, 3, 6, and 11 after the unrelated transplant for graft-versus-host disease (GVHD) prophylaxis. Prophylactic antibiotics were used according to departmental guidelines.
Actuarial rates of OS and progression-free survival (PFS) were estimated by the Kaplan–Meier method. Prognostic factors for survival were evaluated using the Cox proportional hazards model for univariate analyses. Multivariate analysis was not performed due to a small sample size. Statistical significance was defined at the 0.05 level. Analysis was performed using STATA 7.0 software (StataCorp, 2001, Stata Statistical Software, Release 7.0; StataCorp LP, College Station, TX).
Complete remission was defined as the absence of original monoclonal protein in urine and serum by immunofixation, < 5% plasma cells in bone marrow aspiration, and no increase in size or number of lytic bony lesions. Progressive disease was defined as one of the following: 1) > 25% increase in serum or urine monoclonal protein, or plasma cells in the bone marrow, or 2) increase in the size or number of lytic bony lesions.19
Patient characteristics are shown in Table 1. All patients except one in the allogeneic group had secretory myeloma with measurable paraprotein levels in serum or urine. Two patients of 14 in the autologous group and 1 of 26 patients in the allogeneic group underwent HDT in untreated disease recurrences. All the other patients received salvage therapy with conventional regimens (vincristine, doxorubicin, and dexamethasone;, thalidomide; and dexamethasone) before proceeding to transplantation.
Table 1. Characteristics of Patients Who Underwent Salvage Autologous or Allogeneic Transplantation
n = 14
n = 26
TP: transplantation; MUD: matched unrelated donor; Ig: immunoglobulin; BM: bone marrow; PB: peripheral blood; CR: complete remission; MR: minimal response; NR: no response; PD: progressive disease; PR: partial remission; ED: early death, before response evaluation; CR; complete response.
Median age in yrs (range)
Median months between diagnosis and second TP (range)
Median months between first TP and second TP (range)
light chain only
Durie Salmon stage at initial diagnosis
Median number of prior chemotherapy regimens (range)
Response to TP
Chromosomal abnormalities were detected in 11 patients before HDT. Deletion of chromosome 1 was the most common abnormality observed (8 of 11 patients), whereas deletion of chromosome 6 was noted in 3 patients. Two patients with chromosome 1 abnormalities also had concurrent deletion of chromosome 13. All these abnormalities are known to be associated with MM.22 No macroscopic myelodysplasia was reported on bone marrow morphology before salvage HDT.
Survival in the Autologous Group
The median follow-up among survivors was 18 months (range, 2–69 mos). The response rate in the autologous group was 64% (complete response rate of 21% of patients and a partial response rate of 43%). Only 1 of 14 patients (7%) died of nonrecurrent disease causes in the first 100 days. Overall, 2 patients (1 with pneumonia and 1 with sepsis/multiorgan failure) died of nonrecurrent disease causes (14%). The median OS was 29.5 months and the median PFS was 6.8 months (Fig. 1).
Survival in the Allogeneic Group
The median follow-up among survivors was 30 months (range, 13–66 mos) in the allogeneic group. The response rate in the allogeneic group was 69% (complete response rate of 31% and a partial response rate of 38%), Three of 26 patients (11%) died of nonrecurrent disease causes (acute GVHD in 1 patient and sepsis/multiorgan failure in 2 patients) in the first 100 days. Overall, 7 of 26 patients (27%) died of nonrecurrent disease causes (acute GVHD in 1 patients, chronic GVHD in 3 patients, sepsis/multiorgan failure in 2 patients, and secondary malignancy in 1 patient). The median OS was 13 months and the median PFS was 7.3 months (Fig. 2). Grade 2-4 GVHD was noted in 31% of patients, whereas chronic GVHD was noted in 42% patients (limited in 19% and extensive in 23%).
Prognostic indicators for survival in the allogeneic transplant group
On univariate analysis, an interval of > 1 year between the first and the salvage transplant (P = 0.02) predicted a significantly better OS (Table 2) (Fig. 3). Age, cytogenetics, disease status at the time of transplantation, type of donor, tumor mass, β2 microglobulin level, serum albumin level, and chronic GVHD also were studied and were found to have no effect on survival. None of the prognostic factors evaluated for the allogeneic group were found to have a significant impact on survival in the autologous group.
Table 2. Univariate Analysis Showing Factors Found to Have a Favorable Impact on OS after a Salvage Allogeneic Transplantation
HDT with ASCT results in higher complete remission rates, OS, and EFS in patients with MM. However, this approach is not curative and > 70% patients undergoing autografting experience disease recurrence within 3 years.1–3
To our knowledge, there is no standard salvage treatment for patients who develop a diseae recurrence after an autograft. Various treatments ranging from conventional agents to salvage autologous transplants and allogeneic transplants have been used. Newer antimyeloma agents including thalidomide, lenalidomide, bortezomib, and arsenic trioxide have shown encouraging results, with response rates of 30–40% reported in heavily pretreated patients with both recurrent and refractory disease. However, the durability of these responses remains to be assessed.5–11
Both autologous and allogeneic transplants have been performed for salvage therapy after failure of the first autograft. In the series reported to date, both salvage autografts and allografts were associated with response rates in > 50% of patients that were not durable.10, 11 Furthermore, the potential benefits of allogeneic transplants were neutralized by high nonrecurrence mortality.10 These transplants were performed with myeloablative conditioning regimens and were associated with high rates of treatment-related complications. It has been postulated that the use of reduced-intensity, nonablative regimens may lower the incidence of TRM without compromising the graft-versus-myeloma effect.15
We reviewed the outcomes of salvage autografts and reduced-intensity allografts performed at our institution. This is a retrospective analysis with known pitfalls, including a heterogeneous patient population, potential bias in patient selection, use of several different preparative regimens, and a shorter median follow-up after autologous transplantation.
The salvage transplant was safe because the 100-day nonrecurrence mortality rate reported in the autologous group was comparable to that reported with first autografts.2, 3 In contrast, the 100-day nonrecurrence mortality rate in the reduced-intensity allogeneic group (11%) was significantly lower than the rate of 30–50% that has been reported for myeloablative transplants.14 Despite the presence of persistent or progressive disease, two-thirds of these heavily pretreated patients achieved a complete or partial response after salvage high-dose therapy. However, these responses in general were not durable. Disease progression remained a major cause of failure in both autologous and allogeneic groups. The median OS was relatively longer in the autologous group, which could be related in part to a shorter follow-up and fewer patients in this group.
Both autologous and allogeneic transplants can be performed as salvage after failure of an autograft. Both approaches are safe and result in complete plus partial responses in 68% of patients. However, these responses are short-lived and disease progression remains a major cause of treatment failure. Because of limited efficacy, high cost, and treatment-related complications, salvage autologous and allogeneic transplants should be performed only in the setting of a clinical trial, preferably in comparison with conventional nontransplant approaches using newer agents such as bortezomib or lenalidomide.7, 8 Other approaches including vaccines and maintenance therapies may also be studied.23, 24