The Impact of Histologic Grade on Acute Graft Versus Host Disease Response and Outcomes

Acute graft versus host disease (GVHD) is a major cause of morbidity and mortality after allogeneic hematopoietic cell transplantation (HCT). Based on established consensus grading criteria, a correlation between clinical grading and survival has been well established. A histologic grading system for acute GVHD of the skin and gastrointestinal (GI) tract has also been well established, and is routinely used by pathologists when reporting this diagnosis. There are limited data, however, on the correlation of the histologic grade with clinical grade and its ability to predict outcome. We thus undertook this retrospective analysis to further evaluate the impact of histologic grade on treatment response and outcome. Among 503 consecutive patients who underwent allogeneic HCT from 2005-2013, we identified 234 patients with 303 episodes where tissue biopsies (skin and GI) confirmed presence of acute GVHD. There were 140 (46%) GI and 163 (54%) skin biopsies. Histologic grade was analyzed and correlated with clinical grade of GVHD, day 28 treatment response, and outcome.  Although both GI and skin histology were significantly correlated with overall clinical grade (p<0.001), there was a stronger association with GI histologic grade (R=0.559) compared to skin (R=0.315). GI histologic grade was lower than clinical grade in 64% of GVHD episodes, the same in 15%, and higher in 21% of occurrences. In contrast, skin histologic grade was lower than clinical grade in only 4% of GVHD episodes, the same in 17% and higher in 79%. Overall histologic grade, independent of tissue site, and overall clinical grade were both significantly associated with day 28-treatment response. 71% (n=70) of patients with histologic grade 1 GVHD had a complete response (CR) to therapy; whereas only 6% (n=1) with histologic grade 4 GVHD achieved a CR.  In univariable analysis, worst GVHD by histologic grade was predictive of non-relapse (HR 1.67, 95% CI 1.30-2.14, p<0.001) (Figure 1) and overall mortality (HR 1.40, 95% CI 1.15-1.70, p<0.001). Similarly, overall clinical grade was predictive of non-relapse (HR 1.41, 95% CI 1.02-1.93, p=0.036) and overall mortality (HR 1.60, 95% CI 1.29-1.99, p<0.001). In multivariable analysis, overall clinical grade remained the strongest predictor of non-relapse (HR1.41, 95% CI 1.02-1.94, p=0.037) and overall mortality (HR 1.65, 95% CI 1.32-2.06, p<0.001), while histologic grade became non-significant. In conclusion, histologic GVHD grade has some correlation with clinical grading and treatment response, and may play a role in further supporting the severity of GVHD, particularly in specific tissues. Overall clinical grade, however, remains the strongest predictor of mortality post-transplant.