Narrowing the knowledge gap in atypical HUS

In this issue of Blood, Brocklebank et al¹ present the largest single-center cohort with suspected complement-associated atypical hemolytic uremic syndrome (CaHUS) treated with eculizumab. CaHUS is a rare kidney disease in which complement activation occurs on endothelial cell surfaces, causing a thrombotic microangiopathy. The authors provide detailed information of the clinical characteristics at presentation and a complete genetic analysis of these patients from the National Renal Complement Therapeutics Centre (NRCTC) in the United Kingdom.

Patients with CaHUS typically present with thrombocytopenia, microangiopathic hemolytic anemia, and acute kidney injury. Approximately half of individuals have an inherited (complement factor H, complement factor I, CD46, complement factor B, and C3 mutations) or acquired (factor H autoantibodies [FHAAs]) complement abnormality identified. Treatment historically included supportive care with/without plasma exchange, but outcomes were poor, with end-stage kidney disease (ESKD) or death occurring at first presentation in many patients. More recently, small, single-arm trials of eculizumab in CaHUS have been reported; however, the rarity of the disease has made clinical trials difficult. In the United Kingdom, all cases of suspected CaHUS are referred to the NRCTC, and these cases are the subject of this report.

The authors report a significantly higher 5-year ESKD-free survival in eculizumab-treated patients with CaHUS (85.5%) compared with an untreated, genotyped matched cohort (39.5%). While recognizing the limitations of this type of comparison, these data support the efficacy of eculizumab in CaHUS. The study also reports a low relapse rate, which has been reported by other researchers, and clearly argues against the need for lifelong use of this expensive drug.^2,3 

It is important to examine the differences between the eculizumab-treated cohort and the control group. The latter was selected on the basis of the presence of a pathogenic complement genetic variant or FHAAs, thus supporting a diagnosis of CaHUS. Clinical information at presentation for the control group was not available. The eculizumab-treated group consisted of patients referred to the NRCTC, with a suspected diagnosis of CaHUS. Pathogenic variants or FHAAs were detected in only 47% of patients, raising the question of the presence of complement dysregulation in the other 53%.

Although the study provides useful data on the epidemiologic features of CaHUS, its presentation, clinical course, and outcome, and delineates novel causes of “secondary” HUS/thrombotic microangiopathy (TMA), it is evident that much remains to be understood about this rare disease.

The most important outstanding issues raised by this study include the following:

1. The difficulty of diagnosing CaHUS. The treated cohort included 51 patients who were later diagnosed as having secondary thrombotic microangiopathy (positive predictive value, 192/243 = 79%). Conversely, 544 patients were referred for eculizumab therapy, but not treated, when a diagnosis of CaHUS was considered unlikely. After full evaluation, a pathogenic mutation was found in 40 of these patients. Thus, the diagnostic accuracy is limited at best, even in an esteemed center of expertise: with a sensitivity of 69% and a specificity of 90%. Clearly, biomarkers are needed to make an early and accurate diagnosis, and ascertain that patients receive the appropriate therapy. Rapid DNA diagnostics will be an important step forward, but they are not the holy grail. For example, in many patients (41%), no mutation was detected. Outcome in these patients was intermediate (5-year ESKD-free survival, 65%), suggesting that this group is heterogeneous and may include patients with CaHUS and those with other causes of a CaHUS-like condition. The recently proposed ex vivo endothelial complement activation assay, although not specific, might improve sensitivity of our diagnostic algorithms.⁴ 

2. Outcome in patients with CaHUS is not optimal, even in the era of eculizumab. Although eculizumab therapy is effective, many patients in this study had persistent severe chronic kidney disease (CKD), with 46% of patients with CKD stage ≥3 and 21% of patients needing renal replacement therapy. In a multivariate analysis, older age at presentation, lower estimated glomerular filtration rate at presentation, high systolic blood pressure, less severe thrombocytopenia, and a longer interval between diagnosis and start of eculizumab therapy were all associated with persistent kidney injury. It would have been interesting to analyze the association between the initial changes in laboratory parameters after the start of eculizumab therapy and persistent kidney injury. Increasing awareness of CaHUS, and earlier referral, may be a way to improve outcome.

3. Uncertainty of the relevance of variants of uncertain significance (VUSs). Although genetic analysis is pivotal in evaluating patients with suspected CaHUS, the pathogenicity of discovered variants is often unclear. These variants are categorized as VUSs. Because the 5-year ESKD-free survival in treated patients with CaHUS and with VUSs is comparable to the survival of those with pathogenic complement variants (with a similar relapse rate), many VUSs may be pathogenic. Functional analysis of VUSs will add to a better characterization of CaHUS.

4. Uncertainty about the efficacy of eculizumab in non-CaHUS (secondary thrombotic microangiopathies). Complement activation is observed in many patients with secondary TMA. In many patients, this is a physiological process, well regulated by the intrinsic complement inhibitors. Still, ongoing debate exists whether complement inhibition with eculizumab could benefit some or all of these patients. Unfortunately, this study does not provide an answer. By chance, the study included 51 patients treated with eculizumab, who were later shown not to have CaHUS. The 5-year kidney survival in this group was worse (57%), suggesting limited efficacy of eculizumab. However, to draw conclusions, a comparison with untreated controls, matched for secondary TMA diagnosis and the absence of genetic variants, would be needed.

5. Relapse rate after eculizumab withdrawal. This study demonstrated a low relapse rate of 1 per 9.5 patient-years (py) in patients with a pathogenic genetic variant, 1 in 10.8 py in patients with VUSs, and no relapse in 67 py in patients with no variant detected. These data should be interpreted with some caution, as the distribution of patients stopping therapy did not match the overall patient population. For example, of the 14 patients stopping therapy with a pathogenic variant, 9 had a variant in the membrane cofactor protein (MCP) gene, which is associated with overall good outcome, even when untreated. Still, the data add to other recent studies that support considering eculizumab withdrawal in patients with CaHUS, especially in patients with no pathogenic variants.^2,3,5,6 

Brocklebank and colleagues have provided a wealth of data, improving our understanding of CaHUS. Still, many questions remain, and further work is needed to improve the management of patients with TMA and suspected CaHUS.

Conflict-of-interest disclosure: N.C.A.J.v.d.K. has received consultancy and lecture fees from Novartis, Roche, and Alexion. J.F.M.W. has received consultancy fees from Novartis and Alexion.