The HDM/SCT procedure for AL amyloidosis was introduced in 1998 by Comenzo et al., and represented a major breakthrough for AL amyloidosis . Indeed accumulating data have indicated that HDM/SCT can suppress underlying monoclonal plasma cell disease and improve the patient’s quality of life . The purpose of induction chemotherapy is to reduce the light chain load and to improve organ function prior to HDM/SCT. Amyloidosis is characterized by a relatively small tumor mass, so small that induction chemotherapy is often not necessary before proceeding directly to high-dose therapy. To date, the use of induction chemotherapy prior to HDM/SCT has been evaluated in a few studies, which have reported negative results. Pretreatment with two cycles of oral melphalan/prednisone prior to the transplantation did not improve the results of a prospective randomized trial . Similarly, the data from Perz et al.  indicated that administering vincristine, doxorubicin and dexamethasone (VAD) before HDM/SCT did not increase the hematologic response rate. Nonetheless, the outcome of BD induction prior to HDM/SCT has remained unknown.
Several studies have reported the results of adjuvant BD combination with HDM/SCT, and the addition of bortezomib to the HDM used as a conditioning regimen prior to stem-cell transplantation is feasible and well-tolerated by patients with AL amyloidosis . The BD regimen has also been used following HDM/SCT to improve the depth of the response for patients who achieve less than VGPR. Nineteen of 28 patients received post-transplant BD chemotherapy, with 67% of these achieving CR and 60% organ responses . A complementary approach of administering two cycles of BD prior to and as conditioning for HDM/SCT also yielded very good response rates of hematological remission in 9/18 patients after the induction, and all 11 patients evaluable after HDM/SCT attained CR/VGPR . The combination of BD with cyclophosphamide has also shown encouraging response rates in patients with AL amyloidosis, even without high-dose therapy [40, 41]. Additionally, most of the patients in these two series were either transplant ineligible or relapsed. Therefore, based on the current knowledge regarding the use of BD in the treatment of AL amyloidosis, it is clear that the BD regimen is effective and fast acting in patients with AL amyloidosis. Our data also show that BD therapy prior to HDM/SCT can quickly reduce the FLC level and the tumor burden and can furthermore delay the progress of involved organs, thus improving the safety and efficiency of HDM/SCT. Therefore, the BD regimen is an appropriate therapy before HDM/SCT. For patients who have already achieved CR prior to HDM/SCT, it is difficult to assess whether HDM/SCT was necessary, and only a clinical trial can provide an answer. Regardless, we believe that additional therapy with HDM/SCT may be able to improve the depth of the response and further extend PFS of those patients.
Our data demonstrate that BD induction followed by risk-adapted HDM/SCT is a more effective strategy than HDM/SCT alone for treating newly diagnosed patients with AL amyloidosis. The majority of patients (85.7%) in the BD induction group achieved hematological responses, including more than half (67.9%) who achieved CR at one year post-therapy. The BD induction group also displayed better PFS and OS rates than the group that received HDM/SCT alone. HDM/SCT treatment in patients with cardiac involvement remains challenging, particularly for patients with stage III cardiac involvement. In our series, 44.4% (4/9) of the patients with stage III cardiac involvement died during follow-up, but the median survival time was more favorable than previously reported .
The disadvantages of induction chemotherapy in patients with AL amyloidosis include the potential risk of side effects, the risk of further worsening organ function and the consequent delay of HDM/SCT. Our data demonstrate that the majority of patients can tolerate two cycles of BD treatment, and that the toxicity of the BD regimen is moderate and manageable. All toxicities were reversible and did not influence the subsequent HDM/SCT procedure because MEL toxicity differs from that of bortezomib. However, additional attention should be given to some toxicities that can result in bortezomib or dexamethasone dose reduction, including edema, peripheral neuropathy, and thrombocytopenia. Edema is a common toxicity for high-dose dexamethasone, particularly in patients with renal involvement and low serum albumin levels. Because the serum albumin levels of all five patients who developed grade 3 edema were lower than 2 g/dl, support treatment with diuretics and serum albumin is important for these patients. Peripheral neuropathy also needs to be carefully monitored during BD induction, as 18% of the patients developed grade 3 neuropathy in this study, apparently higher than in other studies [24, 25]. One study demonstrated that 1.6 mg/m2 once-weekly bortezomib dosing can reduce toxicity compared to 1.3 mg/m2 twice-weekly dosing . Subcutaneous bortezomib is another option to reduce the incidence of peripheral neuropathy and has been confirmed in a randomized, phase 3 study for multiple myeloma patients  and was also proven in a small series of AL amyloidosis patients . In addition, we observed a high incidence of herpes zoster infection during BD induction; thus, anti-virus prophylaxis appears to be necessary for these patients. In our study, the occurrence of TRM in our study population of 56 patients was 3.6% (2/56), a lower value than in most reports. Moreover, no patients in the BD induction group died because of treatment-related complications. The possible reasons for the relatively low TRM observed include the risk-adapted approach of HDM/SCT, the induction chemotherapy used in half of the patients, the small proportion of patients with severe cardiac involvement and racial differences.
This study has several limitations. First, all of the patients had renal involvement, which most likely constitutes a selection bias because the study was conducted in a nephrology institution. Second, neither group met the median PFS and OS rates, and the long-term patient prognosis requires further study. Third, it is unclear whether the PFS rate of the patients who achieved CR after the BD induction following HDM/SCT is equivalent to the CR achieved with BD therapy alone. Fourth, the sample size of this trial is too small to allow patient stratification based on known risk factors, as well as to allow a subgroup analysis, so the results for stratification or subgroup analysis cannot match the results of other studies. We will continue this study and enroll more patients to make this stratification or subgroup analysis more clear. Additionally, the appropriate salvage regimen for patients who had no response to BD induction following HDM/SCT is also unclear.