Patients with Paroxysmal Nocturnal Hemoglobinuria and Hemolysis Demonstrate More Frequent Disease-Related Features Than Those without Hemolysis, but Similar Proportions Experience Thromboembolism

Introduction: Patients with paroxysmal nocturnal hemoglobinuria (PNH) suffer high morbidity and mortality, however, data characterizing these patients by hemolytic status are unavailable. The objective of this analysis is to describe disease characteristics in patients with PNH by the presence of hemolysis.

Methods: The International PNH Registry is a prospective, observational study of patients with a PNH clone of 0.01-100%. At enrollment, historic data are collected from disease start, prospectively and semi-annually thereafter. Patients enrolled on or before 8 April 2015 were eligible for this analysis and were assessed during two follow-up periods: Cohort A: follow-up time without reported hemolysis, defined as time from disease start until the last reported untreated LDH value <1.5 x upper limit normal (ULN); Cohort B: follow-up time with reported hemolysis, defined as time from LDH ³ 1.5 x ULN post-disease start, to most recent untreated follow-up. Mortality and physician-reported symptoms are only assessed after enrollment in the Registry; only the subsets of Cohorts A and B untreated with eculizumab at enrollment with prospective follow-up time were eligible for analysis of these endpoints. Patients could contribute follow-up time to both cohorts. Demographics and disease characteristics were assessed and compared using the chi-square test or FisherÕs exact test.

Results: A total of n=1,012 patients in Cohort A and n=1,565 patients in Cohort B met the criteria for inclusion; n=191 patients contributed follow-up time to both cohorts. Median years of follow-up time (range) without hemolysis (Cohort A) and with hemolysis (Cohort B) was 2.2 (0.0, 54.2) years and 1.2 (0.0, 37.2) years, respectively. Median disease duration (disease start until last untreated follow-up) was 2.7 (0.0, 54.2) years in Cohort A vs. 5.1 (0.0, 62.9) years in Cohort B. At most recent follow-up, 69.4% of patients in Cohort A and 39.6% in Cohort B had a history of aplastic anemia. Earliest reported granulocyte clone size (median, [range]) without hemolysis (Cohort A) was 4.0% (0.01, 100) and with hemolysis (Cohort B) was 65.6% (0.01, 100.0). The annual rate of thrombotic events (TEs) was higher in Cohort B than in Cohort A (Table). A total of 676 patients in Cohort A and 1,203 patients in Cohort B were untreated at enrollment; 548 and 1,022 of these patients, respectively, had follow-up after enrollment. Median age (range) at enrollment in patients with follow-up time after enrollment was 40.0 years (7, 89) in Cohort A and 43.0 years (8, 90), in Cohort B. Over a median of 1.0 years of prospective follow-up in each cohort (range 0.0, 9.1, and 0.0,9.5), mortality was higher in Cohort B than in Cohort A (1.4% (22 deaths) vs. 0.1% (1 death); p<0.001). Physician-reported symptoms were reported in both cohorts; all except fatigue were significantly higher during hemolysis (Cohort B) (Figure).

Conclusions: This real-world analysis from the International PNH Registry showed that patients with PNH had increased morbidity and mortality after reported hemolysis compared with patients without hemolysis. Thrombosis is the leading cause of mortality in patients with PNH. While the annual rate of TE was higher in hemolytic patients, the proportion of patients with TE was similar in the hemolytic and non-hemolytic cohorts. Despite similar follow-up time, mortality was higher in the hemolytic than non-hemolytic cohort. Symptoms of PNH were reported in both the hemolytic and the non-hemolytic groups; fatigue was the most frequent symptom in both cohorts, and was not statistically different between the two groups. Ongoing monitoring of the development of hemolysis in patients with PNH is important. 

Table. Thrombotic events and mortality in Cohort A (no reported hemolysis) and Cohort B (reported hemolysis)

	Cohort A (n=1,012)	Cohort B  (n=1,565)	P-value
Thrombotic events (TE)
Number of patients with TE, n (%)	50 (4.9)	95 (6.1)	NS
Number of TE events	59	114	--
Rate of TE/year (95%CI)	0.01 (0.01, 0.02)	0.03 (0.02, 0.03)	<0.001
Mortality
Number of patients at risk of death (untreated at enrollment in the Registry)	676	1,203
Number of deaths, n (%)	1 (0.1)	22 (1.8)	<0.001
Rate of death/year (95%CI)	0.001	0.013	.002

CI, confidence interval; NS, not significant

Figure. Presence of physician-reported symptoms by hemolytic status