High dose melphalan in primary systemic amyloidosis: status quo?

Primary systemic light chain (AL) amyloidosis is clonal plasma cell disorder characterized by the deposition of amyloid fibrils in various tissues which are derived from the polymerization of fragments of light chains and result in organ dysfunction. The bone marrow plasmacytosis is usually mild or moderate and the clinical presentation and the outcome of these patients are affected mainly by the extent and the severity of organ involvement by amyloid, particularly of the heart [1]. Patients with AL amyloidosis are characterized by significant heterogeneity in terms of clinical presentation and prognosis. In a rare disease and with such differences in the characteristics and the prognosis of patients, the definition of a treatment as a ‘standard’ is not easy. The randomized trials that have been conducted in patients with AL amyloidosis are limited and of small size. Alkylating agents and standard dose steroids were the mainstay of therapy for several years, however with modest effectiveness: a hematologic response was seen in fewer than 50% and was slow, and complete responses were rare and only a few patients showed significant improvement in organ dysfunction [2]. High dose melphalan (HDM) with autologous stem cell transplant (ASCT) was introduced as a treatment option for patients with AL amyloidosis in the mid 1990s, and although this procedure was associated with significant treatment related mortality (TRM), it induced high rates of complete hematologic responses and organ responses [3]. Pulsed high dose dexamethasone was also associated with improved responses, which occurred rapidly, but is poorly tolerated in patients with AL, especially those with multisystem involvement. Oral melphalan combined with dexamethasone (MDex), in a regimen in which dexamethasone was used for 4 days every month, resulted in hematologic responses in 67%, with almost 30% of patients achieving a complete hematologic response and 48% an organ response [4]. This regimen is effective and relatively non-toxic; however, in patients with cardiac involvement the results are poor [5,6]. A multicenter randomized trial conducted in France compared MDex to HDM in patients with AL amyloidosis [7]. In this study MDex was superior, or at least not inferior, to HDM, but this study was conducted in unselected patients in centers with limited experience, and the TRM in patients treated with HDM was high. Thus, the ‘optimal’ treatment for patients eligible for HDM remained unresolved.

In this issue of Leukemia and Lymphoma, Gertz et al. [8] analyze the outcome of 434 patients with AL treated with HDM in a tertiary, specialized center. This is the largest published series of patients with AL treated with HDM, with a long follow-up; furthermore, the data set includes cardiobiomarkers [9]. A hematologic response was recorded in 76%, including hematologic complete response (CR) in 39% and organ responses in 47%. After a median follow-up for the survivors of about 4 years, the median survival for patients who achieved a CR has not been reached, and for those with a partial response (PR) is about 9 years but just 2.5 years for those who did not respond. Elevated cardiobiomarkers remain a major unfavorable prognostic feature, as well as high levels of free light chains (FLCs). The TRM associated with the procedure was 10%, although 25% of patients were stage III according to cardiobiomarker staging [9]; this relatively low TRM, considering the unfavorable characteristics of patients with AL amyloidosis, indicates the importance of risk stratification and careful selection of patients who will undergo the procedure and of the management of these patients in centers with experience and expertise.

So where are we now? Should high dose chemotherapy (HDT) with ASCT still be considered as first choice treatment, or is MDex the standard of care for patients with AL amyloidosis, even those who are eligible for HDT? Unfortunately we are far from defining a regimen as a standard of care for a complex disease such as AL. HDT is a very effective treatment, but only for a minority of patients with AL: only about 15–30% are eligible during initial diagnosis, depending on the series. Thus, we still need an effective regimen for the majority of patients with AL. We now have reasonably long follow-up to support that for non-transplant eligible patients MDex can provide significant benefit, especially in patients at intermediate or low risk. Novel agents, and especially bortezomib, seem to induce response rates that are comparable to those obtained with HDM, even in patients with relapsed AL [10], while in a small series of patients treated upfront with bortezomib with dexamethasone (BD), CRs were sustained even without any alkylating-agent based consolidation [11]. Thus, the reasonable next step is to compare bortezomib-based regimens to MDex in non-transplant eligible patients. But then what is the role of HDT in patients who may be eligible for this procedure in the era of novel agents? We cannot bypass an effective treatment such as HDM, especially considering the results from studies such as these presented by Gertz et al. in this issue. Could we further improve the outcome after HDT? A reasonable approach would be the addition of a short induction phase with an active agent such as bortezomib, especially in patients with high levels of involved FLCs, in order to rapidly reduce the levels of the toxic light chains. This could result in a significant improvement of the performance of patients before ASCT, while even patients who were initially ineligible could become candidates for ASCT. With response rates of about 70–80% with 4–6 cycles of bortezomib-based therapy, a significant increase in the number of patients who could achieve a CR after HDT could be expected with this strategy. Induction with bortezomib is associated with higher response rates before ASCT, significantly higher rates of CRs and very good partial resonses (VGPRs) after HDM, and improved progression-free survival (PFS) in patients with myeloma [12]. The importance of CR in patients with AL cannot be overemphasized: CR is associated with regression of amyloid deposits [13], and is associated with the highest rates of organ response and a significant improvement in survival compared to PR. An equally important task is to identify those patients who are at high risk for complications due to HDM–ASCT. Cardiac biomarkers but also other parameters such as serum albumin and uric acid may help identify patients for whom the risk exceeds the expected benefit. However, a significant proportion of patients will present with stage III disease: these patients may have a significant benefit from HDM, as indicated by the less steep survival curve following the first year after HDM, although with a significant risk for TRM. For most patients with severe cardiac involvement or of advanced age HDM may not be an option, and MDex, or even novel combinations such as CTD (cyclophosphamide, thalidomide, dexamethasone), perform poorly [5,14]. So, is there no hope for these patients? The novel agents, especially bortezomib, may offer some opportunity; however, for several of these patients, their fate seems to be predetermined, and they succumb to their disease within 2–3 months despite management with the most modern therapies. Thus, along with the introduction of effective therapies, still the therapeutic window may be narrow, and a timely diagnosis with rapid institution of appropriate therapy is important.