Incomplete Complement Inhibition in Patients with PNH on Eculizumab - 5 Year Experience from the National PNH Service Leeds

Background

Paroxysmal Nocturnal Hemoglobinuria (PNH) is a rare acquired disorder characterised by intravascular hemolysis and thrombosis. Patients with symptomatic PNH are commenced on the complement inhibitor eculizumab (600mg weekly for 5 weeks then 900mg 2 weekly). This monoclonal antibody targets C5 in the complement cascade, halting terminal complement activation thus inhibiting intravascular hemolysis. In some patients intravascular hemolysis is not adequately controlled on the standard regimen. Patient symptoms, transfusion requirements and raised Lactate dehydrogenase (LDH) levels are indicators for suboptimal control of PNH and review of eculizumab dosing.

The 50% hemolytic complement (CH50) test is a functional assay assessing capability of serum complement components of the classical pathway to lyse sheep red blood cells pre-coated with rabbit anti-sheep red blood cell antibody. Patients with complement inhibited PNH should demonstrate absent lysis. As the test is expensive and difficult to organise, we tested if incomplete complete blockade as determined by CH50 activity would be better to confirm under-dosing than LDH value.

Methods

The Leeds (UK) PNH National Service reviewed patients who underwent CH50 assay between January 2015 and March 2020. All patients were on eculizumab with clinical concerns regarding suboptimal control of PNH. Patients receive eculizumab infusions intravenously every 14 days and routine follow up from the PNH Service. Serum samples were obtained 24 hours prior to infusions for CH50 assay; LDH values were routinely collected. Complete complement blockade was defined by <10% CH50 activity; intravascular hemolysis was indicated by LDH value >1.5x upper limit of normal (ULN). Confidence intervals were set at 95% and significance set at p<0.05.

Results

In the study period, 327 tests (median 2, range 1 - 8) were carried out in 146 patients (median age 54 years, range 16 - 89; 74 female). 81% (265) were successful; 19% (62) were unsuccessful due to processing errors. Of the successful tests, 74% (197 in 127 patients) indicated complete complement blockade and 26% (68 in 38 patients) indicated incomplete blockade. Of the patients with incomplete blockade, 68% (26) demonstrated complete blockade on repeat testing and 32% (12) had their eculizumab dose increased.

Clinical symptoms of under-dosing in the 12 patients requiring a dose increase included increased transfusion requirements and/or breakthrough hemolysis (7), pregnancy (2; both returned to 900mg post pregnancy) and significant lethargy (3). Of the patients requiring a dose increase, 3 were on 1200mg before 2015; their dose was increased to 1500mg.

Repeat testing was carried out in 10/12 patients after dose increase; 8 indicated complete blockade; 2 patients were incompletely blocked at 1200mg and received a further dose increase to 1500mg. Further testing indicated complete blockade in 1 patient; 1 required a 3rd dose increase to 1800mg due to incomplete blockade and ongoing transfusion requirement.

Corresponding LDH values were analysed; median LDH for the complete blockade group was 1.16xULN (range 0.54 - 2.16) and 1.28xULN (range 0.76 - 2.38) for the incomplete blockade group. LDH values were not significantly higher in the incomplete blockade group compared to the complete blockade group, p=0.08. There was no significant difference in LDH values pre- and post-dose increase, p=0.38 (Figure 1); median pre-dose increase LDH 1.14xULN; median post-dose increase LDH 1.13xULN. Correlation coefficient shows that CH50 activity was positively correlated with LDH value, r(123)=0.18, p=0.04.

Conclusion

We report the effective utilisation of CH50 analysis where there is clinical concern of suboptimal control of PNH. All patients demonstrating hemolytic activity on CH50 assays indicated subsequent complement blockade following increase of eculizumab dose. Increasing eculizumab is costly requiring robust evidence of suboptimal complement inhibition; a positive correlation between CH50 activity and LDH values was shown however this is not sufficient to guide clinical decisions. LDH values of the incomplete blockade group were not significantly higher than those with complete blockade, suggesting the use of LDH values as an assessment of complement inhibition in patients with ongoing symptoms or transfusion requirements is not sufficient to guide eculizumab dose increases.