Year: 2017
Type: article
Study design: Case Report
Abstract: Anti-thymocyte globuline (ATG) is a well-established immunosuppressive treatment used for aplastic anemia (AA) and transplantation. Different brands of ATG are all associated with a broad list of immune reactions ranging from mild infusion reactions to serum sickness and anaphylaxis.1 For this reason, high-dose steroids, antihistamines, and paracetamol are recommended during ATG treatment. However, immune toxicity during ATG treatment remains very common, and its actual mechanism has not been fully elucidated yet. Here, we provide evidence that complement is largely involved in the pathogenic mechanism of ATG-related immune reactions, which may be classified as complement activation related pseudoallergy (CARPA)2 rather than as a classic IgE-mediated allergic response. We report the case of a 24 years old male who received in 2009 the diagnosis of moderate AA associated with paroxysmal nocturnal hemoglobinuria (PNH). This patient two years later started eculizumab due to overt hemolysis; however, for the subsequent worsening of his marrow failure, he underwent two courses of immunosuppression (while remaining on eculizumab), achieving a complete response of his AA.3 In summer 2016, he started to relapse of AA, with mild cytopenia (but Hb drop was >5 g/dL, in absence of detectable intravascular hemolysis) and re-expansion of PNH hematopoiesis (from 51% at time of best response to 83%). For this reason a further course of immunosuppression was given, again without discontinuing eculizumab; ATGAM was used at the standard dose of 40 mg/day for four consecutive days, in combination with cyclosporine. The global toxicity of ATGAM was moderate, with fever and chills the first day of treatment, followed by generalized pain (mostly arthralgia) and purpuric eruption with vasculitic-like lesions of the extremities. After initial resolution, joint edema and pain reappeared with steroid tapering, which consequently was delayed, with full resolution of clinical manifestations. To investigate the pathophysiology of ATGAM-related disorders and a possible involvement of the complement cascade, we serially assessed the consumption of complement fractions (and in particular of its key component C3) for 3 weeks during and after immunouppressive treatment. In comparison to baseline values, we observed a discrete reduction of C3 and C4 plasma levels, which reached the nadir at day +5, and recovered by day +20 (Figure 1A). We took advantage of his concomitant PNH erythrocyte population (i.e., intrinsically lacking surface complement regulators) to further investigate possible complement activation on these cells (Figure 1A). LDH level remained normal over time, consistent with prevention of intravascular hemolysis due to eculizumab. PNH erythrocytes were about 25% before treatment, and 8.77% of them were opsonized by C3 fragments, as commonly seen in PNH patients on eculizumab treatment.4 The proportion of C3+ PNH erythrocytes increased very quickly after ATGAM start, with a fivefold increase (38%) reached at 48 hours from treatment starting. This eventually resulted in some increase of C3-mediated extravascular hemolysis, as documented by a slight decrease of total PNH erythrocyte percentage.5 The proportion of C3-opsonized PNH erythrocytes and other changes came back to pretreatment values within a few days, consistent with a transient complement activation triggered by ATG. Complement activation after ATG treatment. (A) AA/PNH patient on eculizumab: substantial C3 decoration of PNH RBC is observed during ATG treatment (from 8% to 38%), with no intravascular hemolysis (LDH remains normal) and mild C3-mediated extravascular hemolysis (demonstrated by the mild decrease of the % of total PNH RBC). C3 (and C4, not shown) consumption is immediate, and recovered over the following 3–4 weeks, while clinical symptoms progressively disappeared. This boosted complement activation might have contributed to mild hemolysis seen at day +20 and +35 (just before eculizumab administrations), as possible breakthrough due to an excess of C3 convertase displacing C5 from its complex with eculizumab. (B) AA/PNH patient not receiving eculizumab: ATG treatment was associated with intravascular hemolysis, as indicated by decrease of PNH RBC (from 9% to 2%) and increased LDH (from 261 to 371), with minimal C3 decoration of PNH RBC (from 0% to 7%), possibly nonspecific. C3 (and C4, not shown) consumption is immediate, and recovered within 2 weeks. Yellow histograms represent the % of C3+ RBC within the PNH RBC population. The blue line depicts the % of PNH RBC (type II + type III); the red line depicts LDH (IU/L; normal range <250); the green line indicates the C3 plasma level (mg/dL). The purple arrow indicates the administration of eculizumab (which was given at the standard schedule of 900 mg every 14 days). The insets on the top of the graphs depict some flow cytometry stainings of patients' RBC with a polyclonal anti-C3 antibody (FITC-conjugated) and a monoclonal anti-CD59 antibody (PE-conjugated)4 To further support our conclusions, we performed the same investigations in another AA/PNH patient not receiving any anti-complement treatment. This 19 years old female presenting as severe AA received standard immunosuppression with ATGAM and cyclosporine; the treatment was well tolerated, since only a minor rash was observed. As depicted in Figure 1B, here we observed a progressive decrease of PNH erythrocytes (from 9% to 3% within 72 hours), consistent with increased complement activation eventually leading to intravascular hemolysis (as confirmed by LDH rise). This is in agreement with previous observations describing possible hemolytic crisis in AA patients with significant PNH erythrocyte populations.4, 6 Indeed, in absence of therapeutic blockade of the terminal effector complement, PNH erythrocytes could not accumulate C3 fragments on their surface because of immediate hemolysis (even if we also observed the appearance of a few C3-opsonized PNH erythrocytes, which possibly have not reached the threshold needed to arm the terminal complement). The kinetic of PNH erythrocytes disappearance paralleled that of C3 opsonization seen in the eculizumab-treated patient, suggesting that the extent of complement activation was quite comparable in the two patients. Even in this case, there was a substantial consumption of C3 and C4. The recovery of C3 and C4 was faster as compared with the other patient, consistent with a smoother clinical course associated with a self-limiting complement activation. Previous immunological studies on patients developing serum sickness have demonstrated the presence of circulating immune-complexes, skin immune deposits and consumption of C3 and C41, 7, 8. Here, we take advantage of patients harboring PNH erythrocyte populations to assess the actual role of complement in immune toxicity during ATG treatment. Indeed, when the terminal complement is blocked by the anti-C5 eculizumab, opsonization of PNH erythrocytes with C3 fragments is a sensitive read-out of ongoing complement activation.5 In steady-state conditions, this proportion remains quite stable in individual patients,5 since it results from continuous activation of the complement alternative pathway due to the so-called C3 tick-over. However, any external trigger able to activate complement may change this proportion. Our data clearly demonstrate that ATGAM was able to activate the complement cascade through its classical pathway (likely due to immune-complexes, as confirmed by C4 consumption paralleling that of C3), eventually leading to increased C3 opsonization (and increased C3-mediated extravascular hemolysis). Conversely, in absence of eculizumab this complement activation is not detected as surface C3 deposition, but rather as transient increase of hemolysis.4, 6 Acute immune toxicity of ATGAM tends to occur to first exposure (and reduces at subsequent administrations), improves with decreased infusion speed (and total dose), and is self-limiting. This clinical scenario resembles CARPA (and more in general of type III hypersensitivity, mediated by immune-complexes) rather than typical IgE-mediated allergic responses (which usually require sensitization, and tend to worse at further exposure, irrespective of infusion speed and dose). Our observations provide experimental evidence for the pathogenic mechanisms underlying acute immune toxicity of ATG, paving the way for possible future therapeutic intervention once novel complement modulatory agents may become available.9, 10 SM and AMR conceived the study together with PR, who performed the laboratory work. SM, LM, GB, OV, FDG and FG performed the clinical work, under the supervision of FP and AMR. SM, PR and AMR wrote the manuscript, which was critically revised by all the other authors. AMR has received research support from Alexion, Novartis, Alnylam and Rapharma, lecture fees from Alexion, Pfizer and Jazz, and served as member of advisory board for Alexion and Jazz, as well as consultant for Amyndas. The remaining authors have no conflict of interest to disclose.
Source: American Journal of Hematology
DOI: 10.1002/ajh.24945
Language: English
Subfield: Immunology
Field: Immunology and Microbiology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2765880967