Severity of Chronic Gvhd and Personality Determines Quality of Life and the Activity Profile: Results of a Prospective German Multicenter Validation Trial

The NIH staging and response criteria offer for the first time the chance for uniform documentation and evaluation of physical functioning and quality of life (QoL) aspects of chronic graft-versus-host disease (cGVHD). Here we present the results of an interim analysis of a prospective German multicenter study on the NIH staging criteria in cGVHD.

Hundred-thirty-eight patients (median age 46 years, range 19–64) after allogeneic hematopoietic stem cell transplantation (HSCT) for hematologic malignancies were evaluated according to the NIH criteria based cGVHD activity assessment, the Lee Chronic GVHD Symptom-Scale (L-cGVHD-SC), FACT-BMT, Human Activity Profile (HAP), SF36, Berlin Social Support Scale (BSSS), 24 Item Adjective Measure (24–AM), Hospital Anxiety and Depression Scale (HADS), and the NCCN-Distress-Thermometer. Enrolment occurred between day 100 and 1 year after HSCT or in the presence of active cGVHD also at later time points. Follow-up surveys were conducted at 1, 2, 3, 5, 8, 12 and 18 months after baseline survey. At all time points disease status, comorbidities and medication were documented. Additionally the NIH-Symptom-Scale was applied for clinician rating of cGVHD symptoms.

Ninety-one patients had cGVHD (mild n=29, moderate n=38, severe n=24) while 47 patients had no cGVHD. The cGVHD NIH consensus grading correlated with impairment of the activity profile (r=0.29, p<0.01), and the FACT-G total score (r=0.29, p=0.04). Pain highly correlated with impairment of QoL (FACT-G total score) (r=0.71, p<0.01), and impairment of the activity profile (r=0.30, p<0.01) and was associated with depression (r=0.29, p<0.01). In addition, the activity profile correlated with QoL (FACT-G total score) (r=0.41, p<0.01), anxiety (r=0.20, p=0.02), and mental health (r=0.38, p<0.01). Multiple regression analysis revealed that severity of cGVHD (β=0.27, p=0.01), grade of distress measured by the distress thermometer (β=0.28, p<0.01) and extraverted personality (β= −0.27, p<0.01) significantly predicted QoL as measured by the FACT-BMT. In contrast to the FACT-BMT, the SF36 total score was not predicted from severity of cGVHD in a multiple regression analysis; the only factors that predicted the SF36 total score were comorbid diabetes mellitus (β=0.22, p=0.01) and patients age (β=0.26, p=0.01). Moreover, a multiple regression analysis including the FACT-BMT (β= −0.29, p<0.01), L-cGVHD-SC (β= −0.19, p=0.01), and SF36 showed that the SF36 (β=0.12, p=0.1) did not contribute a significant additional information in the prediction of the HAP-maximum activity score. Using a multiple regression model the HAP-maximum activity score correlated inversely with severity of cGVHD (β= −0.25, p=0.015). Male sex (β=0.24, p=0.02) as well a diligent personality (β=0.24, p=0.02) had a significant positive impact on the HAP-maximum activity score.

The results demonstrate, that severity of cGVHD as assessed by the NIH consensus grading correlates with impairment of physical functioning as well as QoL. Gender as well as personality are significant modifiers of the activity profile and QoL. While the FACT-BMT correlated with severity of cGVHD the SF36 failed to detect cGVHD related impairment of QoL and the activity profile in this cohort of patients in a multiple regression analysis.