Complement Biomarkers for Tracking Breakthrough Hemolysis in Paroxysmal Nocturnal Hemoglobinuria (PNH) Patients Treated with Last-Generation Complement Inhibitors

Background After the introduction of complement inhibitors targeting C5 (eculizumab, ravulizumab), the field of PNH has been revolutionized by the development of proximal complement inhibitors targeting C3 (pegcetacoplan), factor D (danicopan) or factor B (iptacopan). Since they are able to prevent C3-mediated extravascular hemolysis in addition to inhibit MAC-mediated intravascular hemolysis, these compounds result in better hematological responses, but breakthrough hemolysis (BTH) emerged as a possible adverse event. CH50, free-C5 and free-eculizumab have been used as assays useful to monitor therapeutic inhibition, but they remain specific for anti-C5 therapies only.

Methods We investigated complement biomarkers searching for their association with clinically meaningful hemolysis. Instead of looking for residual complement activity, we tracked in vivo complement activation through the complement breakdown fragments sC5b-9, Bb, C4d (assessing terminal, alternative and classical pathways, respectively), and erythrocyte-bound C3d. We analyzed serial samples collected from 2 untreated patients who then received danicopan in monotherapy (total samples 83), and from 8 patients initially treated with eculizumab (total samples n=19), who then received iptacopan (total samples 263, combination therapy with eculizumab n=236, and monotherapy n=27). Data were correlated with clinical and laboratory hemolysis (lactate dehydrogenase, LDH).

Results Samples collected from untreated patients highlight massive and broad complement activation, with increased sC5b-9 (2478±2151 ng/mL; n.r. 20-172), increased Bb (4.578±0.494 µg/mL; n.r. 0.060-0.566), but normal C4d (0.036±0.023 µg/mL; n.r. 0.033-0.219), demonstrating that it is the alternative pathway which enables the terminal pathway eventually causing hemolysis in PNH. Patients on the terminal inhibitor eculizumab treatment showed elevated Bb (3.378±2.124 µg/mL), and elevated sC5b-9 (1273±1007 ng/mL), even if this latter seems slightly lower as compared to that of untreated patients. Erythrocyte-bound C3d was detectable in all samples (32.3±19.2% C3d+ erythrocytes, n=56; n.r. <0.1), consistent with the notion that uncontrolled proximal complement enables C3-mediated extravascular hemolysis. Linear correlation between sC5b-9 or Bb and LDH was not observed, even if spikes of sC5b-9 were observed in concomitance with BTH (as demonstrated by rise of LDH). In contrast, the pattern of these complement biomarkers was quite different in patients receiving proximal complement inhibitors, consistent with their capability of intercepting upstream complement activation. On danicopan monotherapy, both Bb (1.874±0.739 µg/mL) and sC5b-9 (695±225 ng/mL) were lower as compared with pre-treatment values, and C3d deposition on erythrocytes was not detectable. Similarly, on iptacopan monotherapy, both Bb (1.815±1.215 µg/mL) and sC5b-9 (557±475 ng/mL) were lower as compared to baseline values, and C3d deposition on erythrocytes became undetectable. When Bb and sC5b-9 were correlated with LDH looking to all aggregate data, we could not find any linear correlation; thus, when tracked in condition of complete complement inhibition, these biomarkers could not predict BTH. Nevertheless, fluctuations and spikes of Bb or sC5b-9 were seen, and even if they were always associated with LDH increase, symptomatic clinical BTH was observed just occasionally, demonstrating that these biomarkers are useful to track subtle, transient leakages of complement inhibition.

Conclusions Systematic investigation of biomarkers of complement terminal, alternative and classical pathways in PNH patients proves that proximal complement inhibitors disable also the terminal complement, eventually leading to prevention of both intravascular and extravascular hemolysis. Since Bb and sC5b-9 are significantly reduced most of the time, they fail in tracking clinically meaningful chronic residual intravascular hemolysis. Nevertheless, Bb and sC5b-9 are elevated in case of in case of acute hemolysis, confirming transient loss of therapeutic inhibition at time of BTH. Since subtle fluctuations of Bb and sC5b-9 can be seen also in absence of clinically meaningful hemolysis, they track low-level residual in vivo complement activation, which may help in understanding the pharmacodynamics of novel complement inhibitors and how to best use them.

Risitano:Samsung: Membership on an entity's Board of Directors or advisory committees; Achillion: Membership on an entity's Board of Directors or advisory committees; Sobi: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Apellis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Roche: Membership on an entity's Board of Directors or advisory committees; Pfizer: Honoraria, Speakers Bureau; Alnylam: Research Funding; Ra Pharma: Research Funding; Alexion: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Amyndas: Consultancy.