A Retrospective Analysis of Outcomes Following Allogeneic Stem Cell Transplantation for Patients from Appalachian Counties

Introduction

Socioeconomic status has been demonstrated to impact not only medical treatment patients receive, but also outcomes after treatment (Hastert, 2015; Hines, 2014; Kim, 2011; Hackley 2005). Prior studies assert that low income areas include patients with a later cancer stage at diagnosis, an older population, lower income households, a higher percentage of Medicaid population, and lower percentage of residents with a higher education (Hastert, 2015; Bradley, 2002; Lin, 2014). Patients from low income areas may have decreased access to healthcare and limited understanding of cancer treatment options. As a result, there may be differences in their medical treatment (Hines, 2014).

The Appalachian Regional Commission (ARC) demonstrated that the Appalachian population in Ohio, Kentucky and Pennsylvania has a high percentage of poverty and lower education status (Vanderpool, 2019). The Appalachian population has more people living in rural environments, higher levels of obesity, and negative cancer beliefs (Vanderpool, 2019). SEER data combined with CIBMTR data demonstrated that patients from socially disadvantaged areas are referred for transplant less often, and data from Virginia shows a regional variation in referral for SCT for acute myeloid leukemia (AML) (Paulson, 2019; Arora, 2018). Our aim in this study was to determine if allogeneic stem cell transplant (ASCT) outcomes differ between Appalachian (AR) and non-Appalachian residents (non-AR).

Methods

A retrospective review of patient records was conducted for 1168 patients who underwent ASCT from 2008-2018 at The Ohio State University Wexner Medical Center. Patients were classified as either AR or non-AR based on zip code according to ARC designation.

We compared the clinical and demographic variables between the patients from Appalachian area versus not, using Fisher exact test or chi-square test for categorical variables and the Wilcoxon rank sum test for the continuous variables. Overall survival (OS) and relapse-free survival (RFS) estimates were calculated by the Kaplan-Meier method and compared using the log-rank test. Cumulative incidence of acute GVHD, chronic GVHD, relapse and non-relapse mortality (NRM) were analyzed using Gray's test and accounting for competing risks, where the competing risks for aGVHD and cGVHD were relapse or death, the competing risk for relapse was death from any cause and the competing risk for NRM was death due to disease.

Results

Out of the 1168 patients included in our study, 887 (75.94%) were non-Appalachian and 291 (24.91%) were Appalachian residents. There was no significant difference in age (p 0.14) or gender (p 0.54) between the two groups. The non-AR group and AR group did have a statically significant difference (p <0.01) in the proportion of White and Black patients (Table 1).

In both groups, the majority of patients were diagnosed with AML/CMML (42.19% non-AR, 40.55% AR). Other diseases represented included MDS/AA, ALL/PLL, CLL, NHL, CML, HD/HOD, MF, MM; there was no statistical significance with regard to disease distribution between the two populations (p 0.68). Disease related factors including performance status (graded by Karnofsky Score), remission status, comorbidity index, were similar between both groups-as were transplant related factors such as conditioning regimen, donor type, tissue type, CD 34 and CD 3 count (Table 2).

BMT related milestones and complications such as days to engraftment, bacteremia, viremia, fungemia, hemorrhagic cystitis, VOD and pulmonary complications were not statistically significant between the two groups (Table 3). Cumulative incidence of those diagnosed with acute and chronic GVHD were not statistically significant between the groups (Graphs 1-2).

Outcomes of non-AR and AR groups were compared; results demonstrated that relapse, relapse free survival, overall survival and non-relapse mortality were not statistically significant (Graphs 3-6).

Conclusion

Our analysis demonstrates that despite several barriers to medical care, AR patients have similar outcomes to non-AR patients after ASCT. As a result, we encourage providers not to view Appalachian residence as an indicator of poorer outcomes. Instead, we recommend supporting and referring Appalachian patients for transplant as aggressively as non-Appalachian patients. This single-institution study should be evaluated with a larger multi-center cohort.