Eculizumab treatment and impaired opsonophagocytic killing of meningococci by whole blood from immunized adults

Abstract

Eculizumab, a humanized anti-complement C5 monoclonal antibody (mAb) for treatment of paroxysmal nocturnal hemoglobinuria (PNH) and atypical hemolytic uremic syndrome, blocks the terminal complement pathway required for serum bactericidal activity (SBA). Because treated patients are at >1000-fold increased risk of meningococcal disease, vaccination is recommended; whether vaccination can protect by opsonophagocytic activity in the absence of SBA is not known. Meningococci were added to anticoagulated blood from 12 healthy adults vaccinated with meningococcal serogroup B and serogroup A, C, W, Y vaccines. Bacterial survival was measured after 3-hour incubation in the presence of eculizumab or control complement factor D inhibitor ACH-4471, which blocks the complement alternative pathway (AP) and is in phase 2 development for treatment of PNH. In the absence of inhibitors, colony formation units (CFUs) per milliliter in blood from all 12 immunized subjects decreased from ∼4000 at time 0 to sterile cultures at 3 hours. In the presence of eculizumab, there was a >22-fold increase in geometric mean CFUs per milliliter (90 596 and 114 683 CFU/mL for serogroup B and C strains, respectively; P < .0001 compared with time 0). In the presence of ACH-4471, there was a >12-fold decrease (23 and 331 CFU/mL, respectively; P < .0001). The lack of meningococci killing by blood containing eculizumab resulted from inhibition of release of C5a, a C5 split product needed for upregulation of phagocytosis. The results provide an explanation for the large number of cases of meningococcal disease in immunized patients being treated with eculizumab and suggest that vaccination may provide better protection against meningococcal disease in patients treated with an AP-specific inhibitor.

Figure 1.

Complement pathways. Model by which the CPs and complement APs mediate SBA or OPA killing of meningococci. Blocking cleavage of C5 or C7 in the terminal complement pathway prevents formation of MAC, which is required for complement-mediated SBA. In the absence of MAC, antimeningococcal antibodies may offer protection by eliciting antibody-Fc–mediated and/or C3b/iC3b complement receptor–mediated OPA. However, complement receptor–mediated phagocytic uptake as well as phagocytic cell chemotaxis is promoted by C5a, a split product of C5. Thus, blockage of release of C5a by eculizumab could abrogate this additional protective mechanism.

Figure 2.

Effect of eculizumab (anti-C5) and ACH-4471 (a complement factor D inhibitor) on complement pathways. (A,C) CP activity measured by CP Wieslab. (B,D) CP activity measured by hemolytic assay. Horizontal dashed lines represent IC₅₀, which was 18.3 µg/mL for eculizumab. (E) AP activity measured by AP Wieslab assay. At a concentration of 1 µM ACH-4471 (0.58 µg/mL), complete inhibition was achieved.

Figure 3.

Effect of complement inhibitors on survival of serogroup B meningococci in plasma or whole blood from an unvaccinated adult subject. At time 0, ∼4750 CFU/mL N meningitidis serogroup B strain H44/76 were added to plasma or whole blood containing no complement inhibitor or inhibitors indicated. CFUs per milliliter were quantified in aliquots of blood or plasma collected at 1 and 3 hours postincubation. (A) Test conditions without adding anticapsular antibody. (B) Test conditions with adding 25 µg/mL of a meningococcal serogroup B anticapsular mAb (SEAM 12). Data are the mean CFU per milliliter (ranges) from replicate determinations. Δ Plasma, plasma heated for 30 minutes at 56°C to inactivate complement. The results were replicated in a second experiment with blood from the same donor. ACH-4471, factor D inhibitor of AP; ECZ, eculizumab.

Figure 4.

Effect of complement inhibitors on survival of meningococci in whole blood of individual vaccinated adults. (A) Serogroup B strain H44/76. (B) Serogroup C strain 4243. At time 0, ∼3600 CFU/mL N meningitidis were added to blood (gray bars). In the absence of a complement inhibitor, blood from all 12 subjects had sterile cultures (blue bars) by 3 hours. In contrast, addition of eculizumab (ECZ; anti-C5, orange bars) completely blocked bacterial killing of both strains. ACH-4471 (complement factor D inhibitor, green bars) had no effect (sterile culture) on killing of the serogroup B strain H44/76 in 7 of 12 subjects, and no effect on killing of the serogroup C 4243 strain in 10 of 12 subjects (see “Results”).

Figure 5.

Effect of complement inhibitors on survival of 4 N meningitidis strains in whole blood from vaccinated adult subjects. Geometric mean CFU per milliliter (95% confidence interval) after incubation of meningococci for 3 hours. (A-B) Data from 12 subjects tested against serogroup B strain H44/65 and serogroup C strain 4243, respectively. (C-D) Serogroup B strains Ohio University and Quebec tested with blood from 5 subjects immunized with MenB-4C. Irrespective of the strain tested, eculizumab blocked killing of bacteria and the geometric mean CFU per milliliter increased significantly at 3 hours compared with time 0 (P < .0001). In contrast, in the presence of the complement factor D inhibitor ACH-4471, the geometric mean CFU per milliliter of blood decreased at 3 hours compared with time 0 (P < .005 for strains H44/76, 4243, and Quebec; P = .13 for Ohio University). The P values for comparisons between the respective geometric mean CFUs per milliliter at 3 hours for the different treatment groups for each strain are provided in “Results.”

Figure 6.

C5aR antagonist augments anti-C7 inhibition of whole-blood killing of meningococci. Whole blood from 4 subjects vaccinated with MenB-4C was tested for killing activity against 2 serogroup B strains, Ohio University and Quebec, in the presence or absence of complement inhibitors. Data are geometric mean CFU per milliliter (ranges) from 2 experiments with blood from subjects 007 and 011, and CFU per milliliter from 1 experiment each for subjects 009 and 010. (A,C) CFU in blood at time 0 and after 3 hours of incubation in the absence of inhibitor. (B,D) CFU per milliliter in blood after 3-hour incubation with C5aR antagonist (red bars) or ant-C7 mAb (yellow bars), or anti-C7 + C5aR antagonist (red hatched yellow bars). The combination of the C5aR antagonist and anti-C7 mAb blocked killing of both strains to a greater extent than the anti-C7 mAb alone (P < .04 by paired t test on log-transformed data).