Primary refractory multiple myeloma: Real-world experience and recommendations

Multiple myeloma (MM) is the second most common hematological malignancy after leukemia, accounting for ~10% of hematological neoplasms.^1–3 Treatment protocols have progressed significantly in the last 10 years, embracing novel drug classes, such as immunomodulatory drugs (IMiD^®) and proteasome inhibitors (PI), and the expanded use of conventional treatments, such as autologous stem cell transplant (auto-HSCT).¹

Novel agents have been shown to have greater efficacy at all stages of the disease than previous systemic, non-transplant treatments.⁴ This is particularly noticeable in response to initial therapy, with a higher portion of patients achieving complete response prior to moving on to auto-HSCT.⁴ However, a small percentage of patients respond poorly to initial therapy, being diagnosed with primary refractory disease. It has been reported that around 10–20% of patients with MM have double-refractory disease, not responding after exposure to PI and IMiD, constituting a particularly aggressive form of the disease.⁵ Not only is there no consensus for the management of patients with double-refractory MM, but it is also not known how failure to initially achieve at least partial response to IMiD and PI affects prognosis.

In this international, multicenter, retrospective study reflecting real-world experience, patients diagnosed with MM refractory to primary induction therapy with both PI and IMiD were found to improve progression-free survival (PFS) substantially when auto-HSCT was given at first treatment failure. The results were recently reported by A. Jurczyszyn et al. in Leukemia & Lymphoma journal.¹

Study Design

• International, retrospective, multicenter, observational, real-world study
• Across 17 centers in South America, Europe, the United States, and Hong Kong
• All data collected from medical records and clinical notes
• Case selection:
  • New diagnosis of MM
  • Treated with a triple regimen including a PI and an IMiD
  • Received at least four cycles
  • Failure to achieve at least a partial response
• Treatment outcomes classified by the International Myeloma Working Group response criteria

Results

A total of 85 patients were identified across 17 centers; the key demographic data are presented in Table 1. After not even achieving a partial response to induction treatment with bortezomib plus dexamethasone and an IMiD (lenalidomide or thalidomide), 11 patients (13%) proceeded directly to auto-HSCT, and the rest received a second-line treatment:

• 65 patients (76.5%) received a PI/IMiD.
• Six patients (7%) received conventional chemotherapy.
• Three patients (3.5%) received a monoclonal antibody.
• 42 patients (49%) underwent auto-HSCT as consolidation following the salvage treatment.

Table 1. Demographic, biochemical, clinical, and treatment data¹

auto-HSCT, autologous hematopoietic stem cell transplantation, GFR, glomerular filtration rate; Ig,  immunoglobulin; LDH, lactate dehydrogenase; R-ISS, Revised International Staging System; VRD, bortezomib + lenalidomide + dexamethasone; VTD, bortezomib + thalidomide + dexamethasone. *Data available from 57 patients. †Data available from 69 patients.
Characteristic	N = 85
Age at myeloma diagnosis, median years (range)	58 (28–80)
Age ≥ 60 years, %	43
Eligible for auto-HSCT, %	56
Monoclonal protein subtype heavy-chain isotype, %
IgG	60
IgA	13
IgM	2
Light chain only	23
Light chain isotype, %
Kappa	36
Lambda	26
Hemoglobin, g/dL (range)	10.2 (3.5–15.3)
Estimated GFR, mL/min (range)	67 (3.0–105.0)
Lytic lesions, %	81
Serum beta-2-microglobulin level, mg/dL (range)	3.9 (1.3–32.3)
Increased serum LDH level, %	43
Bone marrow involvement, % (range)	55 (6.4–100.0)
R-ISS stage*, %
Stage I	25
Stage II	54
Stage III	21
High-risk cytogenetics†, %	26
Induction therapy, %
VTD	61
VRD	39

The overall response rate (ORR) after second-line therapy was 51%. The data were analyzed by different subgroups:

• Patients who had auto-HSCT directly after frontline therapy had a higher ORR than those treated with other regimens as second-line therapy (91% vs 45%; p = 0.004).
• Patients who had auto-HSCT directly after frontline therapy had a higher ORR than those treated with other regimens as second-line therapy followed by auto-HSCT (91% vs 55%; p = 0.028).
• Patients who underwent auto-HSCT as consolidation, with or without salvage therapy, had a better ORR than those who did not (62% vs 31%; p = 0.001).

Overall, the median PFS after second-line therapy was 21.6 months. Of note, this study identified auto-HSCT consolidation as the only significant independent predictor of longer PFS, regardless of age and induction regimen (multivariate Cox analysis, HR 0.24; 95% CI, 0.13–0.45; p < 0.001).

• Median PFS with auto-HSCT vs no auto-HSCT was 30.9 months vs 4.0 months, respectively (p < 0.001).

• This advantage also translated into a significantly greater median overall survival (OS): 46.4 months vs 11.0 months, with or without undergoing auto-HSCT, respectively (p = 0.002).

After a median follow-up of 44.6 months, the median OS was 35.6 months. However, A. Jurczyszyn et al. could not identify an independent prognostic factor for OS in a multivariate Cox analysis.

Conclusion¹

This study demonstrates that auto-HSCT, modern IMiD and PI, and conventional treatments have efficacy in the salvage treatment for MM refractory to induction treatment. Eligible patients with primary double-refractory MM benefit the most from auto-HSCT, primarily when performed immediately after frontline therapy. While auto-HSCT is an independent predictor of PFS, it is not associated with a statistically significant OS, suggesting efficacy for other salvage treatments in later therapy courses. The study also reports up to 30% of responses to second-line therapies, both novel and conventional, in transplant-ineligible patients.

The authors recommend using auto-HSCT after induction regimen or salvage therapy as the first option for treating transplant-eligible patients with primary double-refractory MM based on the significant improvement in PFS, in line with previously reported studies.⁶