Abstract

Background

The introduction of novel agents has significantly improved response rates in multiple myeloma (MM) patients. The exact impact of quality and duration of response on long-term survival, however, remains controversial.

Methods

We retrospectively analyzed MM patients who underwent high-dose chemotherapy and autologous stem cell transplantation (ASCT) as first-line therapy at our center between June 1992 and February 2012. Treatment response was assessed according to IMWG criteria both immediately before and 100 days after ASCT. Overall survival (OS) was calculated from day 100 after first ASCT. In addition, in a landmark analysis OS was calculated starting from 3 years post first ASCT. Log-rank tests were used to assess the impact of response on OS. P-values < 0.05 were regarded as statistically significant.

Results

1155 patients with symptomatic MM (683 males, 472 females) were evaluated for this analysis. Median age was 56 years (y) (range 24 to 74y). 751 received single, 404 tandem ASCT. Maintenance therapy with interferon, thalidomide, or bortezomib was administered after ASCT to 629 patients. 37 proceeded to allogeneic transplantation and were censored at that time. Median OS was 5.2y (range 4.8 to 5.8y).

The achievement of complete response (CR) or near CR (nCR) before ASCT was associated with a superior OS of 7.0y, significantly longer than for those achieving either a partial response (PR) or very good PR (VGPR) (p < 0.01 and p = 0.04, resp.). Evaluating response on day 100 after ASCT, patients in CR/nCR continued to show superior OS in comparison to those in PR (p < 0.01). However, VGPR post-transplant had similar prognostic value to CR/nCR (p = 0.17). Interestingly, patients remaining in stable disease or minimal response after ASCT also had a favorable OS of 6.4y (range 4.8 to 9.7y), potentially indicating a distinct biological entity.

To assess whether timing of response is of prognostic relevance, we compared outcome of patients reaching CR/nCR before ASCT with those of patients achieving CR/nCR only after ASCT. Early and late responders showed a similar OS (7.0y vs 6.4y, p = 0.2) with no significant difference between the two groups.

For the 3-year landmark analysis 709 patients were evaluable. Remarkably, the depth of response before or after ASCT did not exert any significant effect on survival for those patients still alive 3y after ASCT (p = 0.8 and 0.3 resp.). However, the implementation of maintenance therapy after ASCT was associated with a strong survival benefit (OS 5.6 vs 2.9y, p < 0.01).

To assess the importance of response duration, we compared the OS of patients with sustained CR/nCR at 3yrs after ASCT to patients in all other response groups (non-CR/nCR) and patients who had relapsed in the interval (lost CR/nCR and lost non-CR/nCR, respectively). Remarkably, OS was excellent for patients with sustained CR/nCR (10.2y) as well as with sustained non-CR/nCR (7.8y), with no significant difference between both groups. However, patients with lost CR/nCR or lost non-CR/nCR had an unfavorable OS of 2.4 and 2.7y, resp. (fig. 1). Combining the effects of sustained response and application of maintenance therapy showed a significant benefit for maintenance therapy in patients with both sustained and lost response without annihilating the clear difference in survival between the two latter groups (fig. 2).

Conclusion

In this large, retrospective single-center study, achievement of CR/nCR before and after ASCT was prognostic in regard to OS. Yet, for patients evaluable 3y after ASCT, no further significant influence of the quality of response achieved by ASCT was detectable. While patients with sustained CR/nCR as well as those with inferior yet sustained responses showed similar survival, both groups had a markedly better outcome than those patients losing a once obtained response. This effect was independently improved by the implementation of maintenance therapy, indicating that intervention to sustain achieved responses appears to be a key factor for long-term survival.

Fig. 1

OS of patients with sustained vs lost responses (CR/nCR vs no CR/nCR).

Fig. 1

OS of patients with sustained vs lost responses (CR/nCR vs no CR/nCR).

Fig. 2

OS of patients with sustained vs lost responses with and without maintenance therapy.

Fig. 2

OS of patients with sustained vs lost responses with and without maintenance therapy.

Disclosures:

Ho:Sanofi-Aventis: Consultancy, Honoraria, Membership on an entity’s Board of Directors or advisory committees; Genzyme: Consultancy, Honoraria, Membership on an entity’s Board of Directors or advisory committees.

Author notes

*

Asterisk with author names denotes non-ASH members.