Abstract 3949

Acute kidney injury (AKI) is a common but serious sequela of multiple myeloma. Recently, the International Myeloma Working Group (IMWG) introduced a new renal response criteria with complete (CRenal), partial (PRenal) and minimal (MRenal) responses. This study compares the IMWG criteria with other models to determine their correlation with overall survival (OS).

Patients with multiple myeloma and AKI were identified from centers around the world. OS was calculated from the day of AKI. Renal function was assessed by estimated glomerular filtration rate (eGFR) via MDRD method. Hematologic response was the best achieved as assessed by International Uniform Response Criteria for Multiple Myeloma. Only patients with peak serum creatinine (Scr) ≥ 2.0 mg/dl were included. Separate analyses were performed on the newly diagnosed (ND) versus previously treated (PT) patients.

A total of 158 patients were collected from centers in United States, United Kingdom, France and Austria. Fourteen patients were excluded for having peak Scr < 2.0 mg/dl and one for repeated episodes of AKI. Median age of the 143 patients was 64 (34 – 87) years, 58.7% were male and 115 were newly diagnosed. Median Scr was 5.1 (2.0 to 18.6) mg/dl and median eGFR was 10.6 (2.8 – 36.5) ml/min/1.72 m2. Dialysis was required in 49.3%, and 72.9% were dialysis independent at the end of the study.

By IMWG criteria, median OS ranged between 6.0 months (m) in PT patients with no renal response (NRenal) to 23.7m in patients with CRenal, p = 0.34 (Table 1). In ND patients, OS was 31.4m in NRenal vs 73.8m in CRenal (p = 0.08). However, OS was similar between PRenal and CRenal, p = 0.70 (Figure 1). If only patients with a hematologic PR or better were analyzed, the OS range was 6.0m in NRenal to 23.7m in CRenal in PT patients (p = 0.85) and 36.6m in NRenal to 73.8m in CRenal in ND patients, p = 0.73.
Table 1.
Previously TreatedNMedian SurvivalNMedian Survival (>PR)
    CRenal 23.7 m 23.7 m 
    PRenal 13.1 m 12.8 m 
    MRenal 12.0 m 19.2 m 
    NRenal 16 6.0 m 6.0 m 
    p-value  0.34  0.85 
Newly Diagnosed     
    CRenal 22 73.8 m 22 73.8 m 
    PRenal 27 Not Reached 23 Not Reached 
    MRenal 22 65.6 m 19 65.6 m 
    NRenal 43 31.4 m 21 36.6 m 
    p-value  0.08  0.73 
Previously TreatedNMedian SurvivalNMedian Survival (>PR)
    CRenal 23.7 m 23.7 m 
    PRenal 13.1 m 12.8 m 
    MRenal 12.0 m 19.2 m 
    NRenal 16 6.0 m 6.0 m 
    p-value  0.34  0.85 
Newly Diagnosed     
    CRenal 22 73.8 m 22 73.8 m 
    PRenal 27 Not Reached 23 Not Reached 
    MRenal 22 65.6 m 19 65.6 m 
    NRenal 43 31.4 m 21 36.6 m 
    p-value  0.08  0.73 
Figure 1.

Overall survival (OS) of 115 patients evaluated by the IMWG renal response criteria. OS was not significantly different by Log-Rank test between patients with CRenal and NRenal, p = 0.08.

Figure 1.

Overall survival (OS) of 115 patients evaluated by the IMWG renal response criteria. OS was not significantly different by Log-Rank test between patients with CRenal and NRenal, p = 0.08.

Close modal
The best model tested was one that defined renal response (RR) as regaining or maintaining dialysis independence and no response (NR) as dialysis dependence. RR was associated with a superior survival for both PT patients (13.5 m vs 6.0m (NR), p = 0.03) and ND patients (73.8m vs 20.1m (NR), p < 0.001). RR increased with the depth of hematologic response, p = 0.04. RR occurred in 75% of the PT patients with hematologic PR or better and 66.7% in those with less than PR, p = 0.64. In ND patients, RR occurred in 85.7% of those with PR or better and 54.6% of those with PR or less, p = 0.003. This may explain the differences in OS when hematologic response was considered. In PT patients with hematologic PR or better, renal response did not improve OS (21m vs 6m, p = 0.28) but it remained beneficial in the ND patients (73.8m vs 35.1m, p = 0.04, Figure 2). Renal response was not significantly associated with OS for any patient with hematologic response that was VGPR or better. In a multivariate analysis, RR (p = 0.001), ND (p <0.001) and PR or better hematologic response (p = 0.046) were associated with OS.
Figure 2.

In a separate model, 83 newly diagnosed patients with hematologic response of PR or better were evaluated. Patients who were dialysis independent achieved a renal response (RR) and those who remained on dialysis had no response (NR). Patients with RR had a superior overall survival than patients with NR. Renal response was independent of a hematologic response of PR or better.

Figure 2.

In a separate model, 83 newly diagnosed patients with hematologic response of PR or better were evaluated. Patients who were dialysis independent achieved a renal response (RR) and those who remained on dialysis had no response (NR). Patients with RR had a superior overall survival than patients with NR. Renal response was independent of a hematologic response of PR or better.

Close modal

The results of our study suggest renal response is important to OS but the depth of response was not as important. Renal response was most important when the hematologic response was less than a VGPR. This supports renal response as part of the overall response assessment in multiple myeloma patients. Based on our results, reducing the number of response categories and simplifying the current consensus criteria seems justified.

Median survival of the 143 patients evaluated by the IMWG renal response criteria. Overall survival was calculated by Kaplan Meier method. Differences in survival were evaluated by Log-Rank test.

Disclosures:

Ludwig:Mundipharma, Janssen-Cilag: Research Funding, Speakers Bureau. Hutchison:The Binding Site: Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution