Abstract
Long-term survivors in multiple myeloma (MM), described as those surviving >10 years since their diagnosis, are uncommon. There is paucity of data describing this subgroup of patients and how they differ clinically from the rest.
Patients with MM diagnosed from 1998 to 2000 were identified in the National Cancer Data Base (NCDB). We obtained data associated with socio-demographics, type and location of care facility, as well as the use high dose chemotherapy/autologous stem cell transplant (ASCT) as initial treatment option. Four cohorts were created based on overall survival (OS): subgroup 1 (OS: < median); subgroup 2 (OS: median to 2X-median), subgroup 3 (OS: 2X-median to <10 years) and subgroup 4 (OS: >10 years).
There were 27,987 MM patients. The median OS for the whole group was 26.7 months. Among them, 2,196 (7.9%) were long-term survivors. Subgroups 1, 2, and 3 comprised 54.8%, 19.0%, and 18.3% of the remaining patients, respectively. Majority were males (54.3%) with a mean age at diagnosis of 67.2 years (range, 19-90). Compared to the other subgroups (1/2/3), the long-term survivor subgroup had a significantly higher proportion of patients with high educational level (37.8% vs 28.4%/31.6%/33.9%; P < 0.001), high annual household income (41.5% vs 31.0%/34.2%/36.4%; P < 0.001), residence in a metro area (79.2% vs. 77.8%/78.7%/78.3%).; P=0.003), initial treatment at an academic center (46.6% vs 28.1%/34.6%/39.0%; P < 0.001), and had ASCT as part of initial therapy (16.5% vs 2.5%/6.4%/10.9%; P < 0.001). Multivariable analyses showed that younger age, non-Black race, lower educational level, non-Medicare/Medicaid primary payor, treatment at academic centers, and receipt of ASCT as part of initial treatment were significant independent predictors of survival > 10 years. In contrast, sex, ethnicity, type or geographic location of residence, and median annual household income were not significant.
In the US, approximately 1 in 13 MM patients diagnosed in 1998-2000 are long-term survivors. There are disparities in long-term outcomes according to socio-demographic characteristics, type of treatment facility, and receipt of ASCT as part of initial therapy.
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
