Malignant hypertension and thrombotic microangiopathy: complement as a usual suspect

Nephrologists are familiar with malignant hypertension (MHT), as this severe disorder frequently affects the kidney. Their interest has been recently fuelled by accumulating, and sometimes discrepant, data regarding MHT, renal thrombotic microangiopathy (TMA), complement activation in general and particularly in atypical haemolytic uraemic syndrome (aHUS) [1–6]. In a recent series of MHT cases related predominantly to essential hypertension, TMA occurred in only 5% of patients [7]. It is noteworthy that while TMA is a relatively rare complication of MHT in the overall population, MHT is highly prevalent (34–55%) in patients with a clinical diagnosis of aHUS, a form of complement-mediated renal TMA [1, 2]. The current challenge is to dissect the interconnections between several players (Figure 1): MHT, a disorder with heterogeneous underlying causes; complement activation, a ubiquitous, usually self-limited and not necessarily harmful process; TMA, a complex entity with variable pathological and, to some extent, clinical features; and aHUS, an ultrarare disease that nevertheless requires a targeted therapeutic inhibition of complement. The ultimate goal is to identify those MHT patients who would benefit from a therapeutic complement blockade. (A) The interconnections between MHT, TMA and aHUS are complex. MHT may complicate the course of a range of underlying causes, predominantly primary essential hypertension. In a minority of cases (∼5%), MHT can be associated with renal TMA, the latter being defined in reported series by a variable combination of clinical and pathological features. One of the current challenges is to identify among patients with MHT a minority with complement-mediated TMA, i.e. aHUS, which requires urgent therapeutic complement blockade. Furthermore, MHT may occur during various phases of aHUS: acute phase with overt thrombocytopenia and mechanical haemolysis; chronic phase with chronic kidney disease, severe hypertension but no haematological features; or a subacute phase with renal and haematological features of variable severity (termed ‘subacute aHUS’). The role of complement dysregulation is established in aHUS. It remains to be assessed in patients with MHT related to other causes, including essential hypertension. (B) Different approaches to study of the interactions between MHT and aHUS in previously published series. Distinct clinical and/or pathological criteria were used for patient selection. This explains in large part the discrepancy in the frequency of detected complement gene variants between series. Similarly, different terms refer to patients with coexisting TMA and MHT, including aHUS, ‘hypertension-associated TMA’ or ‘malignant nephrosclerosis’. Data are from refs. [1–4]. The report by Zhang et al., published in this issue of NDT, is a timely contribution to this interesting topic. The authors studied 50 patients with MHT, biopsy-proven renal TMA and a history of pre-existent chronic kidney disease. They found that plasma and urinary levels of several components of the complement cascade (factor B, factor D, properdin, C3a and C5a) were increased and those of factor H decreased in patients with ‘malignant nephrosclerosis’ as compared with controls. Urinary but not plasma complement component levels correlated with the severity of microvascular renal endothelial cell damage but not with the presence of thrombi. Similarly, positive C5b-9 staining was detected along arterioles and capillary walls in 45 (90%) kidney biopsies from patients with MHT, which was not observed in patients with benign hypertensive nephrosclerosis or controls. The C5b-9 staining that was similar to C3a, C5a and C4d staining also correlated with the severity of microvascular endothelial cell injury but not thrombosis. Interestingly, MBL staining in kidney biopsies from these patients was negative. Finally, among 20 tested patients, 5 were found to carry rare variants in complement genes. The authors concluded that dysregulation of the complement alternative pathway plays a role in the activity, severity and outcome of ‘malignant nephrosclerosis’ and that screening for complement defects in patients with this disorder is warranted for the potential use of therapeutic complement blockade. The study has strengths, notably the pertinence of the topic and the assessment of complement activation markers in urine, an approach that deserves wider use in the field of TMA. It also carries limitations. The choice of complement activation markers is intriguing, with the absence of C5b-9 level assessment, particularly in urine, C5b-9 being one of most studied complement markers in various types of TMA. Furthermore, an increase in the plasma level of the fragment Bb would more accurately reflect alternative pathway activation than an increase in factor B (as used by the authors). There is also large variability in the serum and urinary levels of complement components within the cohort of patients with malignant nephrosclerosis and a significant overlap between these patients and controls. Also, the proportion of malignant nephrosclerosis patients who have an increased level of one or several complement components is not available. Results of kidney immunohistochemistry staining for complement component, particularly C5b-9, are to be interpreted with caution as complement deposition may simply reflect trapping of the complement molecules in damaged tissue and does not necessarily result from ongoing or recent complement activation [8]. Finally, detailed clinical and biological features (platelet count, markers of haemolysis, kinetics of serum creatinine, timing of kidney biopsy after MHT diagnosis) are crucially lacking, particularly for the five patients who ultimately were found to carry rare complement gene variants. Nevertheless, this study shows that in patients with ‘malignant nephrosclerosis’, associated or not with constitutional complement dysregulation, C5b-9 deposition occurs along microvascular renal endothelial cells and correlates with the cellular injury thereof. However, this finding is hardly surprising. In the setting of MHT, shear stress leads to the activation and ultimately apoptosis of oedematous and detached endothelial cells. Apoptosis is a well-recognized potent activator of the complement cascade at the cell surface, predominantly promoting activation of the complement classical pathway. This explains the C4d deposition documented by Zhang et al. in the kidney biopsies of included patients. A similar C4d and C5b-9 vascular deposition has been previously documented in scleroderma renal crisis [9], another prototype of MHT-induced renal vascular damage, as well as in antibody-mediated renal allograft rejection [10]. Despite these limitations, the study by Zhang et al. provides some thoughts about the interactions between MHT, TMA and aHUS. The first relates to the identification of those MHT patients with complement-mediated TMA requiring a targeted therapy. The documentation of intrarenal (and to some extent systemic) complement activation in a vast majority of patients with malignant nephrosclerosis included in Zhang et al.’s study precludes the use of these markers to identify a small minority presenting with MHT as a complication of complement-mediated TMA. Complement genetics may provide helpful clues, as was the case for 25% of tested patients from Zhang et al.’s study. Nevertheless, the interpretation of the results of complement genes screening requires careful analysis of the clinical relevance of identified variants. For instance, the hotspot variant in the MCP gene reported by Zhang et al. in their patients with malignant nephrosclerosis is also prevalent in healthy Chinese individuals and is not found in patients with aHUS of a similar Asian origin [11]. Finally, genetics results are not available in an emergency setting. Thus the cost-effectiveness of systematic genetic screening in all patients with MHT is far from being clear. Recently developed in vitro tests of C5b-9 deposition on endothelial cells may prove helpful in this setting but require further assessment and standardization before clinical implementation [12, 13]. The second input of this study proceeds from the same evidence provided by Zhang et al. If intrarenal complement activation occurs in a vast majority of MHT patients, would (transient) complement inhibition be beneficial, regardless of the underlying cause of MHT, and of the presence or absence of complement gene rare variants? The study by Zhang et al. does not provide any clear answer to this question. Moreover, one cannot exclude that complement activation may also abate rapidly after the restoration of normal blood pressure in a majority of MHT cases. Finally, the study of Zhang et al. raises the issue of the classification of patients presenting with MHT and in whom genetic complement dysregulation is subsequently documented. Should they be classified as having an initially unrecognized aHUS, as we would be inclined to suggest, or as having ‘complement-mediated malignant nephrosclerosis’ or ‘hypertension-associated TMA’? Semantics are crucial in the field of TMA and MHT and discrepancies in previously published studies and the present one by Zhang et al. proceed from different approaches and thus distinct populations selected for the analysis (Figure 1). When discussing MHT and TMA, it is time we use a common language. Furthermore, with potent complement inhibitors being increasingly available, complement has become the usual suspect, if not the ideal culprit, in a wide range of renal diseases. However, conviction requires hard evidence. And we are still gathering evidence for the implication of complement in MHT. F.F. has received consultancy and/or speaker honoraria from Roche, Alexion, Apellis, Achillion, Novartis and Alnylam. S.S. has no conflicts of interest to declare. V.F.-B. has received fees from Alexion Pharmaceuticals, Roche and Apellis for invited lectures and/or board membership and is the recipient of a research grant from Alexion Pharmaceuticals. (See related article by Zhang et al. Association between thrombotic microangiopathy and activated alternative complement pathway in malignant nephrosclerosis. Nephrol Dial Transplant 2021; 36: 1222–1233)