Uncontrolled Complement Activation and the Resulting Chronic Hemolysis As Measured by LDH Serum Level At Diagnosis As Predictor of Thrombotic Complications and Mortality in a Large Cohort of Patients with Paroxysmal Nocturnal Hemoglobinuria (PNH),

Paroxysmal nocturnal hemoglobinuria (PNH) is a progressive, systemic, and life-threatening disease characterised by chronic uncontrolled terminal complement activation and hemolysis. Uncontrolled complement activation leads to red blood cell (RBC) hemolysis, platelet activation and subsequently thromboembolism (TE), renal and other organ impairment, pain, severe fatigue, poor quality of life and early mortality. Lactate dehydrogenase (LDH) serum level of ≥1.5x the upper limit of normal (LDH ≥1.5x) is a marker of uncontrolled complement activation that has been used in multinational PNH clinical trials. TE is the leading cause of death in PNH and carries a significant mortality risk, accounting for 46% of deaths in Korean PNH patients. The aim of this clinical research project is to evaluate whether uncontrolled complement activation as measured by LDH ≥1.5x at diagnosis is a suitable predictor of TE and mortality.

A retrospective analysis of the national PNH data registry (301 patients) in South Korea was performed. 224 patients who had reported LDH level at diagnosis were analyzed.

PNH patients with LDH ≥1.5x at diagnosis had a 4.8-fold greater mortality rate compared to the age and gender matched general population (AGMGP; P<0.001). In contrast, patients with LDH <1.5x had a similar mortality rate as the AGMGP (P=0.824). Other factors that could be associated with increased mortality (age, gender or coexistence of aplastic anemia/other bone marrow disorders) did not differ significantly between the LDH ≥1.5x and LDH <1.5x populations of PNH patients.

Using logistic regression analysis, LDH ≥1.5x at diagnosis was significantly associated with premature mortality compared to LDH <1.5x (univariate odds ratio 5.0; 95% CI (1.15, 21.70); p=0.009). In a multivariate logistic regression with mortality as the response variable, which contained LDH ≥1.5x at diagnosis, age, gender, and aplastic anemia/other bone marrow disorders as explanatory variables, LDH ≥1.5x at diagnosis was shown to be an independent predictor of mortality (OR=10.57, 95% CI: (1.36, 81.93), P=0.024).

We performed a sensitivity analysis which identified that, unlike the LDH 1.5x threshold, consideration of an LDH threshold higher than 1.5x was not a significant predictor of premature mortality compared to the population of patients with LDH less than these respective higher thresholds: LDH ≥3.0x (OR 1.8; 95% CI (0.78, 4.09); p=0.162), and ≥5.0x (OR 2.0; 95% CI (0.91, 4.32); p=0.082).

Furthermore, using LDH ≥1.5x at diagnosis as the threshold detected 96% of patients with TE, which is one of the multiple severe clinical complications in PNH and itself a risk factor for premature mortality. We performed a sensitivity analysis which identified that, unlike the LDH 1.5x threshold, consideration of an LDH threshold higher than 1.5x, similar to the lack of significant predictor of premature mortality, also missed approximately 50% of the life-threatening TE population compared to the 1.5x LDH threshold (only 67% detection with LDH ≥3.0x and only 47% with LDH ≥5x).

This evidence demonstrates that uncontrolled complement activation as measured by LDH ≥1.5x the upper limit of normal at diagnosis is a strong and independent predictor of clinical complications and mortality in PNH independent of age, gender or the coexistence of aplastic anemia/other bone marrow disorders. LDH ≥1.5x clearly identifies a population of PNH patients with a high risk of life-threatening complications and premature mortality (4.8-fold) from the remaining population of PNH patients that has a survival rate similar to AGMGP. Applying higher thresholds of LDH at diagnosis does not identify a population at significant risk for premature mortality and also does not further concentrate patients at risk, but on the contrary, would miss more than 50% of PNH patients at significant risk for life-threatening complications and consequences. Thus, early therapeutic intervention in PNH patients with LDH ≥1.5x ULN at diagnosis is imperative to prevent TE and other life-threatening complications caused by uncontrolled complement activation and hemolysis, such as kidney disease and pulmonary hypertension, as well as premature mortality.

Lee:Novartis: Honoraria; Alexion: Honoraria. Chung:Novartis: Research Funding.