Clinical Features and Survival Outcomes in Diabetic Patients with Newly Diagnosed Multiple Myeloma (NDMM) Enrolled in the Connect® MM Registry

Background:

Multiple myeloma (MM) is typically diagnosed in patients (pts) aged 65-74 y (SEER: Myeloma. 2019); therefore, many pts present with concomitant metabolic disorders such as diabetes (18%-22% prevalence of type 2 diabetes [T2D; BMC Cancer. 2020;20:489]). Obesity (body mass index [BMI], ≥ 30 kg/m² ) is associated with increased prevalence of T2D (Int J Clin Pract. 2007;61:737), and both are suggested risk factors for MM (Blood. 2012;119:4845; Bone Marrow Transpl. 2014;49:1009). Results from single-center, retrospective studies have shown poorer clinical features and survival outcomes in diabetic pts with MM vs their nondiabetic counterparts (Eur J Haematol. 2012;89:320, Br J Cancer. 2014;111:628). The Connect^® MM Registry (NCT01081028) is a large, US, multicenter, prospective observational cohort study of largely community-based pts with NDMM. Reported here is a descriptive analysis assessing differences in baseline (BL) characteristics, treatment patterns, and survival outcomes in diabetic vs nondiabetic pts enrolled in the Connect MM Registry.

Methods:

Pts with NDMM (N = 3011) were enrolled at 250 community, academic, and government sites. Eligible pts were aged ≥ 18 y with symptomatic MM diagnosed ≤ 2 months (mo) pre-enrollment per International Myeloma Working Group criteria (Leukemia. 2009;23:3). Diabetic pts were defined as those with a history of diabetes and also receiving antidiabetic treatment at BL; those without a history of diabetes and not receiving antidiabetic treatment at BL were defined as nondiabetic pts. Descriptive statistics were used for BL characteristics and treatment regimens. Survival outcomes (progression-free survival [PFS] and overall survival [OS]) were analyzed using Cox regression and adjusted for covariates.

Results:

Of the 2845 pts included in this analysis, 467 (16%) were diabetic, and 2378 (84%) were nondiabetic. The median age was 69 y for diabetic pts and 66 y for nondiabetic pts, with fewer Black pts (12.2% vs 19.5%) in the nondiabetic group. The proportion of men was higher in the diabetic vs the nondiabetic group (61.9% vs 56.6%). Similar percentages of pts had ECOG PS of 0-1 in both groups (53.3% vs 55.3%). Obesity was more prevalent in diabetic vs nondiabetic pts (55.9% vs 30.4%). More diabetic pts had a history of hypertension requiring treatment vs nondiabetic pts (84.2% vs 52.5%). The prevalence of chromosomal abnormalities was comparable between the groups (within the limitation that data were missing in a substantial number of pts). Among MM-defining criteria, diabetic pts had a higher incidence of hypercalcemia (11.8% vs 8.5%), anemia (50.3% vs 46.5%), and increased creatinine (23.8% vs 19%), whereas measured kidney function was similar. A lower proportion of diabetic than nondiabetic pts received stem cell transplantation (SCT; 28.9% vs 41.5%) and triplet regimens during first induction (48.2% vs 52.8%%). The use of ≥ 2 novel agents during first line (1L) was lower in diabetic than nondiabetic pts (27.8% vs 34.6%), but use of a single novel agent (63.4% vs 60%) and alkylator in 1L (20.6% vs 22.4%) was comparable. The top five 1L regimens were similar for both groups, with lenalidomide + bortezomib + dexamethasone being the most common 1L regimen, and lenalidomide the most common maintenance therapy. Diabetic pts had significantly shorter adjusted median PFS (24.4 vs 31.5 mo; hazard ratio [HR], 1.24; P = 0.001; Fig. 1A) and median OS (60 vs 72.6 mo; HR, 1.29; P = 0.001; Fig. 1B) than nondiabetic pts. In a subgroup analysis by BMI, diabetic pts had significantly shorter OS than nondiabetic pts (for BMI < 30 [50.8 vs 73.1 mo; HR, 1.51; P = 0.0001] and BMI ≥ 30 [56.4 vs 78.6 mo; HR, 1.51; P < 0.0001]).

Conclusions:

Diabetic pts with NDMM had poorer BL features and were less likely to receive SCT and triplet regimens and/or aggressive upfront treatment. Regardless of BMI, diabetic pts had shorter median OS than nondiabetic pts, suggesting that diabetes might be a stronger factor for survival than obesity. These results support a previous analysis of the Connect MM Registry that identified history of diabetes as a predictive prognostic factor for poor OS (Br J Haematol. 2019;187:602). Limitations of this analysis include limited detail on the care/management of diabetes and steroid use/dosage (which can be challenging for glycemic control). These findings highlight the need to provide better supportive care for diabetes management in MM pts to improve survival.