The first trial to show the benefit of high dose therapy in multiple myeloma was the IFM90 trial, comparing conventional dose chemotherapy with high dose therapy (TBI 8 Gy and melphalan 140 mg/m2) and autologous-BMT that found significantly increased response rate, event-free survival and overall survival with high dose therapy.r However, the complete response rate was only 22% and 5 year event-free survival from diagnosis was 28%. The benefit was most marked for patients who achieved complete response or very good partial response with a 5 year survival of 72%.
The use of the higher dose of melphalan (200 mg/m2 ) dose is primarily based upon a 1995 prospective, randomized French cooperative study, that compared melphalan 140 mg/m2 plus 8 Gy total body irradiation versus melphalan 200 mg/m2 in 282 newly diagnosed, previously untreated, symptomatic patients less than 65 years of age.r One hundred and forty two patients were treated with melphalan (200 mg/m2) versus 140 patients treated with melphalan (140 mg/m2) plus TBI (8 Gy). Comparison of high dose melphalan 200 mg/m2 with the TBI/melphalan 140 mg/m2 demonstrated faster haematologic recovery, less transfusion requirements, shorter hospitalizations, and a lower incidence of severe mucositis (30% versus 50%). While the median duration of event-free survival was similar in both arms (21 months), survival at 45 months was significantly better in patients receiving melphalan 200 mg/m2 (65% versus 45 %).r
Another multicentre studyrby a UK group, compared standard conventional-dose combination chemotherapy to high-dose therapy (melphalan 200 mg/m2) and ASCT in persons less than 65 years of age and found significantly higher complete response rate in the ASCT group than in the standard-therapy group (44% versus 8%). The rates of partial response were similar (42% and 40%). As compared with standard therapy, ASCT increased median survival by almost 1 year (54.1 months versus 42.3 months). There was also a trend toward a greater survival benefit in the group of patients with a poor prognosis as defined by a high beta-2-microglobulin level. It was concluded that high-dose therapy with autologous stem-cell rescue is an effective first-line treatment for patients with multiple myeloma and who are younger than 65 years of age. Other studies using higher doses of melphalan or the addition of alkylating agents have not demonstrated significant improvements either in terms of response rates or in overall survival.
Similarly, other randomized studies, comparing melphalan 200 mg/m2 with more intensive preparative regimens, such as thiotepa, busulfan, and cyclophosphamider or high-dose idarubicin, cyclophosphamide,r did not result in better outcomes than melphalan. A recent randomised study comparing standard melphalan 200mg/m2 with melphalan 280mg/m2 with amifostine protection resulted in a higher CR + nCR rate in the escalated melphalan dose range (22% V 39%) but without demonstrable benefit in PFS nor OS. Amifostine was effective in mitigating grade 3 and 4 GI toxicities in the melphalan 280mg/m2 group.r
While the RCTs performed by the French (IFM) and English (MRC) groups demonstrated a survival advantage for auto-BMT over standard therapy, other studies have confirmed a benefit of auto-BMT in terms of response rate and EFS, but did not find any evidence of increased overall survival, and recent meta-analyses have not found evidence of a survival advantage for HDT as compared to standard therapy.rrrrr
Of note is the phase III study of the Italian group comparing HD melphalan (200mg/m2) with intermediate dose (100mg/m2). 298 patients were randomised to either the 200mg/ m2 group or the 100mg/m2 group. All underwent double autograft. Overall response rates were 79 % and 72 % respectively. Median PFS was superior in the 200mg/m2 group (31.4 v 26.1 p= 0.01) but with no difference in OS.r
As such autologous-BMT with melphalan 200 mg/m2 conditioning continues to be regarded as standard of care in patients aged under 65 diagnosed with multiple myeloma. It remains to be shown whether this will remain the case as the use of multi-agent chemotherapy regimens incorporating biological agent's increases and as more highly efficient new drugs become available. At this stage, however, available evidence suggests that auto-SCT upgrades the response to novel agent- (e.g. bortezomib/lenalidomide) containing induction regimens, and that auto-SCT and novel agents are complementary rather than alternative. This has been confirmed with the publication of the results of a multinational randomised study comparing consolidation utilising cyclophosphamide - dexamethasone- lenalidomide with melphalan 200mg/m2 and ASCT. Groups were further randomised to receive maintenance post consolidation with lenalidomide and prednisone or lenalidomide alone. The group receiving mel 200mg/m2 and ASCT had a significantly better median PFS (43.2 months v 28.3 months) and at 3 years, a significantly better OS (86% v 73%). Lenalidomide maintenance had no effect on either PFS nor OS at 3 years.r A conceptually similar phase III study evaluating consolidation with oral melphalan, prednisone and lenalidomide versus HD melphalan 200mg/m2 as well as post consolidation maintenance therapy with or without lenalidomide has been published. Once again HD melphalan and auto SCT appears to be a superior consolidative intervention with demonstrably better PFS and OS (median PFS 43 v 22.4 months; OS at 4 yrs. 86.1 % v 65.3%). Moreover in this trial, lenalidomide maintenance was associated with improved PFS but not OS.r
While tandem melphalan-conditioned auto-BMT has been demonstrated to confer a survival advantage in studies by the French and Italian groups, this was only in patients who had a VGPR with the first transplant.rr The results of these studies are also made more difficult to interpret in light of the recent introduction of novel agents as part of induction, consolidation and/or maintenance therapy in myeloma. It has been established, for example, that a similar benefit to tandem transplantation can be obtained with thalidomide as maintenance post auto-BMT.r
Long term follow up of the German single v tandem ASCT study confirms overall non inferiority of single ASCT v tandem ASCT for both PFS and OS. Again tandem ASCT improved CR and n CR response rates c.f. single ASCT.r
HD melphalan and auto-SCT has also been evaluated in the group of patients with myeloma and dialysis dependent end stage renal failure. A large retrospective analysis from MD Anderson of patients with renal failure, dialysis dependence and myeloma undergoing HD melphalan and auto-SCT found all grade organ toxicities were higher in the melphalan 200mg/m2 group c.f. 140mg/m2. As such the lower melphalan dosing schedule may be preferred in this group.r
Primary Systemic Amyloidosis
Auto-BMT is associated with a high response rate in patients with primary systemic amyloidosis (AL), followed by a significant amelioration of organ damage in most responding patients. Unfortunately, auto-SCT is associated with a high toxicity, reflecting underlying organ damage secondary to amyloid deposition.
Conditioning regimens in AL are based on adaptations of high-dose melphalan. The usual melphalan dose is 200mg/m2 although a ‘risk-adapted’ approach, with dose reduction to 140mg/m2 used in higher-risk patients in order to reduce transplant-related toxicity.r Criteria for defining ‘high-risk’ patients are; age >60, increased creatinine level, performance status (PS) 2 or compensated cardiac failure. Reduced dose melphalan has been associated with a decreased response rate in some studiesr but not in others.r
Overall, high-dose melphalan results in significant responses (VGPR and CR) in 50-60% of patients, with complete responses (defined by a negative immunofixation and normal free immunoglobulin light chain) in about one third of patients.
Median time to achieve a response is 3-4 months, although organ responses (which follow haematological responses in most patients) can take up to 2 years. Unlike most series in patients with myeloma, the overall relapse rate is low, especially among patients who achieve a CR after auto-BMT. In the Boston series reported by Skinner et al, the relapse rate at 10-years post-BMT in patients who achieved a CR was 21%.r
Together, the high response rates and low relapse rates in patients with AL who survive auto-BMT translate into long-term (3-5 year) survival rates of 45-60% in most large series. (Table 1)
Unfortunately, auto-BMT with melphalan conditioning is also associated with a high risk of mortality and morbidity in patients with AL, with most series demonstrating a TRM of 11-43%. (Table 1) Cardiac arrhythmias, cardiac failure, cardiogenic shock, gastrointestinal bleeding and infection are the most frequent complications causing early TRM, while renal insufficiency and renal failure develop in up to 20% of those who survive auto-BMT.
Several factors have been shown to be predictive of TRM and transplant outcome in AL including;
- cardiac involvement (as indicated by clinical features – congestive cardiac failure, thickened intraventricular wall on cardiac echocardiography, and/or elevated cardiac biomarkers – cardiac troponins cTnT and cTnI, and pro-brain natriuretic peptide NT-proBNP)r
- renal failurer
- involvement of more than two organsr
- higher pre-transplant levels of serum free light chains (FLC)r
- degree of response achieved after transplantrr
- poor performance statusr
Longer term follow up of various AL amyloidosis cohorts suggest improving TRM in more recent time periods with overall improving post transplant survival. Transplantation in specialised centres performing a critical number of autografts per annum (>4) and higher dose melphalan conditioning appears to be other important factors in reducing TRM and long term survival.r
Table 1: Outcome of patients with primary amyloidosis undergoing autologous transplant
||No of patients
||Overall response rate (CR) %
|| TRM (%)
|| Overall survival (%)
| Boston (US)r
||13 (100 day)
||60 (3 year)
| CIBMTR (multicentre)r
18 (30 day)
27 ( 1 year)
66 (1 year)
56 (3 year)
| UK (multicentre)r
||23 (100 day)
||50 (5 year)
| Mayo Clinicr
||11 (100 day)
||60 (5 year)
|French Intergroup (MAG-IFM)r
||21 (100 day)
||45 (3 year)
||7.5 (100 day)
||60;43;30 (@ 5, 10, 15 years)
While long-term survival appears to be prolonged after auto-BMT, especially in patients who achieve a CR after transplant, there are few studies that have directly compared auto-BMT with standard dose chemotherapy, and the results of these are conflicting. Recently the 20 year updated experience of the Boston group has been published confirming the importance of a haematologic CR in determining long term survival.r
A case-control study conducted by the Mayo Clinic, which compared auto-BMT with standard therapy, showed better survival in the intensively treated group, with a 4-year survival of 71% vs 41%.r However this study was a non-randomised retrospective review and many of the patients in the ‘control’ arm received only melphalan and prednisolone. In contrast, a recent RCT comparing high-dose melphalan with standard dose melphalan and dexamethasone showed a better outcome in patients not randomised to auto-BMT when analysed on an intention-to-treat basis.r However the results of this study must be interpreted with caution as it included only 50 patients in each arm and only two-thirds of those randomised to HDT actually underwent auto-BMT. Indeed, when the analysis was restricted to those who actually received the assigned therapy there was no difference found between the two arms.
The benefit offered by auto-BMT and the population who benefits most from auto-BMT therefore still remains to be clarified. Moreover, it remains unclear what difference treatment with new agents, including bortezomib and the immunomodulators (thalidomide and lenalidomide) will make to outcomes of patients with AL amyloidosis.
© Blood 2002
Figure 1: IFM95 Comparison of 200 mg/m2 melphalan vs. 8 Gy TBI with melphalan 140 mg/m2 as conditioning regimens: overall survival according to treatment arm. While the median duration of event-free survival was similar in both arms (21 months), survival at 45 months was significantly better in patients receiving melphalan 200 mg/m2, 66% versus 46 % (P= 0.05).
Figure 2: IFM95: overall survival after relapse according to treatment arm. There was a trend toward a better outcome for patients treated with 200 mg/m2 melphalan compared with 8 Gy TBI plus melphalan 140 mg/m2 as conditioning regimes. Post relapse treatment was not standardised. Half the patients in arm B received a second cycle of intensive therapy with autologous haematopoietic cell transplantation compared with only 25% in arm A.r
Figure 3: PFS and OS in patients undergoing autoSCT vs MPR consolidation and lenalidomide maintenance v no maintenancer
© New England Journal of Medicine 2014
Figure 4: A) Overall survival in 629 AL amyloidosis patients receiving HD melphalan and autografting. B) OS according to CR versus no CR
© Blood 2002
Transplant related toxicityr
© Blood 2002