Results of the phase III EMN02/HO95 study in patients with newly diagnosed multiple myeloma

After more than 30 years from its introduction, autologous hematopoietic stem cell transplantation (autoHSCT) is still the standard of care for young and elderly fit patients with newly diagnosed multiple myeloma (NDMM).¹

However, in the last decade, the introduction of novel agents able to improve rates and depth of response up to values reported with conventional chemotherapy plus autoHSCT, led to questioning the role of upfront autoHSCT in NDMM. Moreover, further studies re-evaluating the role of consolidation therapy with new drugs were required.

The EMN02/HO95 trial was designed to answer both questions: first, the efficacy and safety of high-dose melphalan (M) plus autoHSCT compared with bortezomib (V) plus M and prednisone (P; VMP) as intensification therapy; and second, the value of adding or omitting consolidation therapy with V, lenalidomide (R) and dexamethasone (D; VRD) in patients with NDMM initially randomized to either VMP or autoHSCT. The results of the study with longer follow-up have recently been published in Lancet Haematology by Michele Cavo and colleagues. Here we report a summary of these results.²

Study design²

This randomized, open-label, phase III study (NCT01208766) enrolled 1,503 patients with previously untreated MM at 172 centers of the European Myeloma Network.

Eligibility criteria

• Age 18–65 years
• Symptomatic MM stage 1–3 according to the International Staging System
• Measurable disease
• World Health Organization performance status Grade 0–2

Randomization

Of the 1,503 patients, 1,493 were eligible for induction therapy, of these 1,354 received induction and were eligible for stem-cell mobilization. After three or four 3-week cycles of V, cyclophosphamide (C), and D induction therapy, 1,197 patients with adequate number of stem cells collected (≥ 4 × 10⁶ CD34+ cells/kg) and with < Grade 3 peripheral neuropathy underwent first randomization:

• 1:1 randomization to either VMP or autoHSCT after high-dose M

or

• at sites with a double HSCT policy, 1:1:1 randomization to VMP, single, or double autoHSCT

Within two months from autoHSCT or the last dose of VMP, 878 patients eligible for the second randomization, were assigned to VRD consolidation group (n = 447) or no consolidation group (n = 431), followed by R maintenance until progression in both groups.

Treatment

• Intensification therapy:
  • VMP for up to four 6-week cycles
    • V, 1.3 mg/m² intravenously (IV) or subcutaneously on Days 1, 4, 8, 11, 22, 25, 29, and 32
    • M, 9 mg/m² orally on Days 1–4
    • P, 60 mg/m² orally on Days 1–4
  • AutoHSCT after high-dose M
    • M, 200 mg/m² IV in patients with creatinine clearance ≥ 40 mL/min, and 100 mg/m² in patients with creatinine clearance 15–40 mL/min. Patients receiving double autoHSCT, two courses of high-dose M were administered 2–3 months apart
• Consolidation therapy:
  • VRD for two 28-days cycles
    • V, 1.3 mg/m² IV or subcutaneously on Days 1, 4, 8, and 11
    • R, 25 mg orally on Days 1–21
    • D, 20 mg orally on Days 1, 2, 4, 5, 8, 9, 11, and 12
• Maintenance therapy, 28-day cycles, until progression or undue toxicity:
  • R, 10 mg orally on Days 1–21

Outcomes

The two primary outcomes, analyzed in the intention-to-treat (ITT) population, were progression-free survival (PFS) from the first (PFS₁) and second randomizations (PFS₂). This article reports the final results of PFS₁ and the second prespecified interim analysis of PFS₂. The safety analyses included all patients receiving at least one dose of study drugs.

Patient characteristics

In the ITT population, demographic and baseline clinical characteristics were well balanced among the treatment groups (Table 1).

Table 1. Baseline demographics and disease characteristics in the ITT population

Results²

Efficacy

First randomization

At data cutoff date (November 26^th, 2018) the median follow-up was 60.5 months (interquartile range [IQR] 59.2–61.7) in the autoHSCT group and 59.4 months (58–61.8) in the VMP group. A better PFS was observed in the autoHSCT group vs. the VMP group, and it was maintained even in subgroups of patients with poor prognosis. A significant improvement in overall survival (OS) from the first randomization (OS₁) was observed in the autoHSCT group vs. the VMP group in patients with a high-risk cytogenetic profile (in particular patients carrying del[17p]) but not in the overall population. Patient outcomes are reported in Table 2.

Table 2. Patient outcomes from the first randomization

Second randomization

At data cutoff date (January 18^th, 2018), after a median follow-up of 42.1 months (IQR 32.3–49.2), in the VRD consolidation group (n = 447) vs. no consolidation group (n = 431):

• PFS₂, 58.9 (95% confidence interval [CI], 54.0–not estimable) vs. 45.5 (39.5–58.4) months (hazard ratio [HR] 0.77, 95% CI 0.63–0.95; p = 0.014)
• 5-year OS from the second randomization (OS₂), 77.2% (95% CI 68.7–83.6) vs. 72.2% (95% CI 59.3–81.7; HR 0.99, 0.71–1.39; p = 0.96)

Single vs. double autoHSCT

In patients assigned to double (n = 210) vs. single autoHSCT (n = 492), in the ITT population:

• 5-year PFS, 53.5% (95% CI 46.6–61.3) vs. 44.9% (38–53; HR 0.74, 95% CI 0.56–0.98; p = 0.036)
• 5-year OS, 80.3% (74.5–86.4) vs. 72.6% (66.5–79.3; HR 0.62, 95% CI 0.41–0.93; p = 0.022)
• The benefit of double autoHSCT over single autoHSCT was higher in patients with one or more high-risk cytogenetic abnormalities

Maintenance therapy

Maintenance therapy with R was prescribed in 599 patients of the autoHSCT group and 376 of the VMP group:

• Median duration of therapy, 34 months (IQR 13.3–50.8)
• Median PFS from start of maintenance, 50.4 months (95% CI 45.8–57.7) in the overall population, and 58 months (49.1–not estimable) in the autoHSCT group vs. 43.2 months (38.7–50.1) in the VMP group (HR 0.76, 0.64–0.91; p = 0.0030)
• The most frequent reasons for maintenance discontinuation were progressive disease (63%) and treatment-emergent adverse events (28%)

Safety from first randomization

Most frequent Grade ≥ 3 adverse events (AEs) observed in autoHSCT group (n = 652) vs. VMP group (n = 472), included:

• Neutropenia, 79% vs. 29%
• Thrombocytopenia 83% vs. 16%
• Gastrointestinal disorders 12% vs. 5%
• Mucositis 16% vs. none
• Infections 30% vs. 4%

The total number of serious AEs were 368 (66%) vs. 189 (34%) in the autoHSCT vs. the VMP group, respectively. The most common serious AEs were infection, 56% of 368 in the autoHSCT group vs. 37% of 189 in the VMP group. Overall, Grade ³ 3 AEs were significantly more frequent in the autoHSCT group (p < 0.0001), although 58% of patients treated with VMP experienced an AE that led to dose modifications.

A total of 311 patients died from first randomization, 38 (12%) of which were treatment related: 26 (68%) in the autoHSCT group and 12 (32%) in the VMP group.

Conclusions²

The improved PFS₁ observed with autoHSCT compared with VMP was maintained in patients with predicted unfavorable outcomes, supporting the value of upfront autoHSCT even in the era of novel agents for myeloma therapy. No significant differences were observed between the autoHSCT group and the VMP group in 5-year OS₁, probably because of the short follow-up.

Significantly better survival outcomes were observed with double autoHSCT in comparison with single autoHSCT, particularly in patients with del(17p), but, because of the small sample size of some subgroups of patients, these results require further confirmation.

Compared with no consolidation, VRD consolidation therapy significantly improved PFS₂, but not OS₂. However, the final analysis of the second randomization will be reported with a longer follow-up.

Finally, the observed benefit and tolerability of R as maintenance therapy until disease progression or undue toxicity was consistent with others phase III clinical trials.