Crystals or His(stones): Rethinking AKI in Tumor Lysis Syndrome

Tumor lysis syndrome (TLS) represents a significant oncologic emergency often observed after treatment of hematologic malignancies by the release of intracellular contents into the blood.¹ Not surprisingly, when AKI develops in TLS patients, there is a significant increase risk of mortality.^1,2 Several excellent reviews have been written on the pathophysiology of the AKI in the setting of TLS, which largely highlight the hypothesis that AKI results from deposition of precipitated uric acid or calcium phosphate crystals in tubules.^1,3

However, most forms of AKI are initiated by alterations in renal blood flow. Increased vascular tone may lead to renal hypoxia and the upregulation of proinflammatory molecules on the endothelial cell surface, contributing to the release of cytokines and chemokines and acceleration of local injury.⁴ Micropuncture studies led by Conger et al. demonstrated rats given exogenous uric acid and a urate oxidase inhibitor (to prevent conversion of uric acid to allantoin) showed reduced GFR and increased tubular pressures. However, increased peritubular capillary resistance was also observed, suggesting that a vascular component may also contribute to renal injury.^3,5

Unlike most mammals, humans lack the urate oxidase enzyme that catalyzes the conversion of uric acid to allantoin, which has much greater solubility than uric acid. The development of a recombinant urate oxidase (e.g., rasburicase) is now one of several strategies employed in patients with TLS to reduce uric acid.³ It should be noted that clinical trials using urate oxidase have not demonstrated benefits in terms of prevention of AKI or mortality.^1,3,6 On the basis of the increased frequency of urate oxidase treatment in chemotherapy, underlying hypotheses related to TLS-induced AKI need to be reconsidered.

To that end, in the current of issue of JASN, Arnaud et al. provide important evidence suggesting a novel mechanism of renal dysfunction independent of uric acid/crystal formation.⁷ Data were obtained from two cohorts of patients who presented to the intensive care unit with TLS. Despite use of hydration and urate oxidase treatment on the large majority of patients, nearly 90% developed AKI. However, only 3 of 55 patients (who were not treated with urate oxidase) showed evidence of urate crystals in urine, suggesting that other mediators of kidney injury must be involved.

An exciting aspect of this study was the development of a novel model of TLS-AKI to investigate potential mechanisms. The model is based on the inoculation of mice with AML cells (driven by the MML-AF9 fusion gene) and subsequent chemotherapy treatment (doxorubicin and cytarabine), which reduced leukocyte burden but increased serum lactate dehydrogenase, phosphorus, and markers of renal dysfunction (creatinine and BUN). The increased creatinine and BUN levels, though significantly elevated, were lower than levels frequently observed in more aggressive models of AKI, such as ischemia reperfusion. Interestingly, although tubular damage was relatively mild, the authors reported significant impairments in peritubular capillary endothelial structure by EM and loss of barrier function by intravital microscopy. These effects were not due to chemotherapeutic agents directly because the impairment was not observed in the absence of tumor burden.

From this, the authors propose the most novel aspect of the study: that released histone proteins may underlie some of these pathophysiologic effects of TLS. Indeed, extracellular histones are well known to have an effect on vascular function,⁸ and have been shown to activate renal inflammatory pathways via Toll-like receptors (TLR) in different models of kidney disease, including AKI from sepsis.^9,10 The authors reported that histone proteins were elevated in serum of TLS patients and correlated with patient outcomes. In addition, histones were elevated in mice with experimentally induced TLS, whereas direct infusion of histones increased peritubular capillary permeability and decreased expression of the endothelial marker MECA-32. Using endothelial cells grown in culture, the authors confirmed that proinjury and proinflammatory pathways were activated in response to histone exposure. The authors further investigated the potential link of TLRs on endothelial injury and showed that antibodies against TLR4, but not TLR2, blocked histone mediated endothelial cell injury in vitro, and that TLR4 knockout mice were protected from impairments in renal vascular function in response to histone infusion.

These data also suggest that histones may be a reasonable therapeutic target in the setting of AKI associated with TLS. To that end, the authors point out that negatively charged heparin has been shown to neutralize positively charged histones in vitro. The authors therefore used nonanticoagulant heparin and demonstrated that alterations in vascular permeability were significantly attenuated in the mouse model of TLS-AKI.

Overall, the study represents a significant advance toward a better understanding of the pathophysiology of AKI. First, it provides strong evidence that the previous focus on crystal formation is not sufficient to explain TLS-associated AKI. Although this has long been suspected, the current report provides the first evidence of a specific alternate pathway (i.e., histone>TLR4), unrelated to uric acid, with the potential to mediate AKI in TLS. Furthermore, data implicating histone effects via alteration in vascular function represent a potential link to previously described mechanisms in other models of AKI.⁴ Second, the use of nonanticoagulant heparin to sequester liberated histones represents an interesting experimental approach to investigate this pathway. As histones have been suggested to mediate injury in other settings,^9,10 this approach should garner further consideration in other models.

Despite these exciting results, several questions remain. For example, although nonanticoagulant heparin preserved vascular structure and function, the manifestation of AKI was not completely reversed by this treatment. It is possible that other factors released from lysed cells play an additional contributory role to the activation of the kidney injury, and this possibility should be investigated further. In addition, the levels of histones in TLS patients are lower than concentrations used in culture studies for endothelial injury. Whether histone induced endothelial injury is achieved at lower histone concentrations in combination with additional factors in the TLS milieu deserves further consideration. Despite these concerns, the current study provides an important starting point to investigate further and reshape our understanding of this disorder.

Disclosures

D.P. Basile reports employment by Eli Lilly (Spouse), research funding from Calcimedica, and patents or royalties from Vascugen.

Funding

None.

Author Contributions

D.P. Basile was responsible for conceptualization, wrote the original draft, and reviewed and edited the manuscript.