Abstract
Hepatic sinusoidal obstruction syndrome (SOS) is a rare fatal clinical entity seen following hematopoietic stem cell transplant (HSCT). It is more commonly reported to occur following allogeneic HSCT compared to autologous HSCT. Historically, it is known as hepatitis following HSCT. It is thought that endothelial damage to the hepatic venules leading to occlusion of the terminal hepatic venules and hepatic sinusoids is the trigger for the development of SOS. Several risk factors have been associated with this condition. Some of these risk factors are patient related while others are transplant process related. Given the high mortality of this condition, early identification of high-risk patients with severe disease is of utmost importance. The management of SOS varies depending on the severity of the disease. Mild to moderate disease has a good outcome with supportive measures alone, while severe presentation of the disease requires a more aggressive management. Defibrotide is the only Food and Drug Administration-approved therapy and it is reserved for severe cases of SOS. The role of defibrotide as a prophylactic therapy remains under investigation.
Sinusoidal obstruction syndrome (SOS), previously referred to as hepatic veno-occlusive disease, is a rare life-threatening condition. Despite its rarity, it is still amongst the most fatal complications following autologous stem cell transplant ASCT (1). The typical presentation is characterized by right upper quadrant pain, hepatomegaly, jaundice and ascites. It can develop at any time following hematopoietic stem cell transplant (HSCT) but classically during the first 3 weeks. The incidence varies between 0 to 60% in different studies. Children with SOS tend to have higher mortality rate compared to adults (1).
This entity occurs secondary to the release of toxic metabolites during the conditioning phase and direct toxicity from chemotherapy and radiation leading to an injury to the sinusoidal cells and hepatocytes in zone 3 of the hepatic acinus. This phenomenon of endothelial cell (EC) activation creates a prothrombic state leading to the occlusion of hepatic venules and sinusoids (2-4). This review is limited to SOS in adult patients and summarizes the pathophysiology, risk factors, and diagnostic criteria of this condition, as well as discussing therapeutic advances.
Pathophysiology
The main underlying pathophysiology of SOS is damage to sinusoidal ECs. There are direct causes (e.g. irradiation, chemotherapy and conditioning regimen) and indirect causes (e.g. pre-existing tissue injury, immunosuppressive agent and allogenicity) that play a crucial role in EC injury. Under normal conditions, hepatocytes metabolize drugs such as cyclophosphamide via the cytochrome P-450 enzymatic system to toxic metabolites. Glutathione (GSH) enzymatic activity helps transform the toxic metabolites into non-toxic ones, ensuring protection of the liver. Given the poor concentration of GSH and high abundance of P-450 in the centrilobular areas of the liver, there is increased susceptibly to toxic metabolites and higher incidence of SOS commonly in zone 3 of the hepatic acinus of the centrilobular area (3-6). Figure 1 illustrates this phenomenon.
In HSCT, patients receive high doses of toxic drugs during the conditioning regimen (e.g. cyclophosphamide and busulfan). This causes initial EC damage leading to graft versus host disease (GVHD), SOS, capillary leak syndrome, engraftment syndrome and diffuse alveolar hemorrhage (2, 4). The damaged ECs secrete cytokines that cause weakness in the mucosal barrier between the ECs, promoting the escape of red blood cells, white blood cells and platelets between the hepatocytes and sinusoid ECs leading to dissection of this layer and the initiation of the inflammatory process (7, 8). In addition, there is an increased release and expression of von Willebrand factor, intercellular adhesion molecule-1, vascular adhesion molecule-1, plasminogen activator inhibitor-1 (PAI1), and thrombomodulin as shown in Figure 2. This pro-inflammatory process promotes a pro-thrombotic state, especially in allogeneic HSCT (9). The cascade of actions and interactions, as well as the activation of the healing mechanism yielding to fibrosis, leads to the obstruction of normal blood flow and increases venous resistance, causing the development of high portal blood pressure, further liver dysfunction and fluid retention in the form of ascites (9-11).
SOS is reported to develop more frequently in allogeneic HSCT compared with autologous HSCT. In allogeneic HSCT, the acute and chronic GVHD are promoted by both humoral and cell-mediated (T-cell) immunity, which may occur in the small bile ducts of the liver, causing injury and damage to these cells, contributing to the elevation of portal system blood pressure and EC damage (9). This is thought to be related to the conditioning regimen and GVHD prophylaxis used during HSCT. Despite the similar immunological background of GVHD and SOS, PAI1 is found to be elevated in SOS but not in GVHD (12, 13). PAI1 might be used in the future as a marker to distinguish between SOS and GVHD in challenging patients. A list of circulating markers in SOS and GVHD are summarized in Table I.
Clinical Presentation and Diagnosis
Traditionally, SOS was diagnosed according to either the Baltimore or the Seattle criteria as shown in Table II (1). The clinical presentation is related to fluid retention secondary to portal hypertension, manifesting by ascites, painful hepatomegaly, jaundice, encephalopathy, confusion and weight gain. As a result of fluid overload, many organs may be affected, leading to hepatic–pulmonary syndrome, hepatorenal syndrome, and cardiac failure, which will eventually progress to multiple-organ failure with associated high mortality (9, 14).
With the evolving understanding of this entity and with the advancement in the strategy of conditioning regimens, the European Group For Blood and Marrow Transplant developed revised diagnostic criteria, highlighting late-onset SOS. SOS is no longer limited to the 21-day period following transplant. In addition, the typical presence of hyperbilirubinemia is no longer required to make the diagnosis (6, 15). A new grading system has been also incorporated classifying cases as mild, moderate, severe, and very severe (6) as outlined in Tables III, IV and V.
Laboratory findings. The laboratory findings are consistent of and not limited to an increase in conjugated bilirubin and transaminases, prolonged prothrombin time, low albumin level and thrombocytopenia, with rapid consumption of transfused platelets (16, 17). Given the non-specific laboratory and clinical findings of SOS, other diagnoses should be ruled out such as hyper-acute hepatic GVHD, infection (viral, fungal), cholestasis of sepsis, drug toxicity, biliary obstruction and heart failure (mainly right side) (10, 17).
Biomarkers. The use of biomarkers has not been validated as part of the diagnostic criteria of SOS. As previously described above, SOS is associated with an increased expression and release of circulating markers as shown in Table I. Although PAI1 was reported by two studies as a useful diagnostic and prognostic biomarker (12, 18), to date, the use of this biomarker has not been integrated into the diagnostic or prognostic tools. More studies are needed to validate the use of biomarkers in the diagnosis of SOS.
Imaging. Imaging modalities such as Doppler ultrasound, computed tomography, and magnetic resonance can be helpful in narrowing the differential diagnosis, especially Doppler ultrasound. There are no sensitive or specific diagnostic findings in SOS. The imaging studies can be helpful to assess the severity of the disease and treatment response by checking the portal vein flow, waveforms, size of liver, and ascites (10, 16, 19).
Biopsy. Due to the high complication rate of transjugular liver biopsy, it is usually reserved for suspected unclear cases of SOS (14, 17). In the early stages of the disease, the histological features of SOS consist of dilation of sinusoids with engorgement due to red blood cells escaping through the space of Disse. This is associated with hepatocyte necrosis in the perivenular area of the liver as illustrated in Figure 2 (3). At a later stage of SOS, liver biopsy reveals incomplete to complete fibrous obliteration of the lumina of central venules with centrilobular area fibrosis that is due to fibrinogen deposition as part of the healing mechanism (20, 21).
Risk Factors
Knowing the risk factors of SOS is critical to help identify the susceptible patients and help clinicians take measures in order to reduce the incidence of this fatal condition (10, 22). Several studies have looked into predisposing factors associated with the development of SOS. The risk factors can be divided into pre-transplantation and transplantation factors. Table VI provides a detailed list of these factors.
Despite epidemiological studies showing similar incidence of SOS in pediatric and adult populations, the risk of SOS seems to be higher among pediatric patients, particularly those under the age of 7 years. In addition, older patients with low-performance status have higher incidence of SOS (23, 24).
Most of the studies confirmed that a pre-existing liver condition poses a significant risk, where the risk of developing SOS is three to 10 times greater in patients with an elevated serum aspartate aminotransferase AST level, and greater AST elevation correlates with a higher incidence of severe SOS regardless of underlying liver disease (25, 26).
The type and severity of the underlying disease determines the likelihood of developing SOS, with more aggressive disease posing an increased risk (26).
Increased risk for SOS is also seen in patients with underlying genetic predisposition such as those with glutathione-S-transferase M1 (GSTM1-null) genotype, C282Y hemochromatosis allele, and methyl hydrofolate reducatse (MTHFR)677CC/1298CC haplotype (27-29).
Other risk factors are related to the transplant process such as the type of conditioning regimen and the source of graft. The risk is higher in patients receiving matched unrelated and non T-cell-depleted grafts (10, 30-36). Patients who have undergone SCT and patients who received abdominal radiation prior to HSCT are at higher risk of SOS (24, 26). Using high-dose total body irradiation of more than 12 Gy with cyclophosphamide as part of myeloablative therapy relatively increases the risk for SOS (32-34). It was found that delay of transplantation for more than 12 months after diagnosis may also increase the risk for SOS (32, 33).
Many drugs have been implicated in the development of SOS, Prophylactic GVHD regimens using cyclosporine A, sirolimus, and methotrexate have been reported to increase the risk for SOS as these drugs promote EC damage (28, 34, 37). The risk of SOS with the use of busulfan is dependent on the drug pharmacodynamics. It has been shown that the risk is lower with the use of intravenous busulfan with pharmacodynamic monitoring. SOS developed in six out of 18 patients with an initial area under the curve of greater than 1,500 nM min/l compared with only one out of 33 patients with a lower area under the curve (38).
Despite the great benefit of targeted therapy, with an overall lower toxicity profile, whereas bevacizumab may have a protective effect against SOS, the risk for SOS is greatest with the other humanized monoclonal antibody to CD33 gemtuzumab ozogamicin (GO) and the humanized monoclonal antibody to CD22 inotuzumab ozogamicin (INO) (39, 40). In postmarketing data, the incidence of SOS and fatal hepatotoxicity increased with the use of GO, which led to its removal from the market in 2010 (40-44). Nevertheless, GO has been back in market after regaining approval for use in patients with relapsed/refractory acute myeloid leukemia or as frontline therapy in patients with CD33-positive acute myeloid leukemia. The risk for SOS after INO therapy is increased compared with conventional chemotherapy according to a phase 3 trial (45).
Prevention and Treatment
SOS is associated with high mortality and morbidity. Although many of the factors predisposing to SOS are difficult to control, physicians should optimize the patient's status including adequate hydration and medication reconciliation. Closer monitoring should be offered for high-risk patients. Addressing underlying hepatitis and match compatibilities prior to transplant, and adjusting the conditioning regimen are ways to help reduce SOS risk.
The treatment varies according to the severity of the disease. Mild disease is usually self-limited and requires no treatment. Supportive care is the first intervention offered to provide symptomatic relief by managing the volume overload and the associated electrolyte disturbances. Seventy-five to 80% of cases of SOS are reversible with the use of supportive measures alone (17). However, there is still an unmet need to provide a better outcome in patients with severe disease. Several preventive options and treatment strategies have been explored.
Prevention
Ursodeoxycholic acid (UDCA). Several studies have looked into the efficacy of UDCA in preventing SOS and hepatic GVHD, although the results were not consistent in all of the studies. The largest randomized study conducted by Rutuu et al. (46) failed to show a reduction in the incidence of SOS, the other two studies showed a reduction in both the incidence of SOS and grade 3 acute hepatic GVHD and better 1-year survival (47, 48). As result, UDCA has been included in the transplant guidelines to be used as prophylactic agent for SOS in patients undergoing allogeneic HSCT. The suggested doses of UDCA are either a total daily dose of 12 mg/kg or fixed dose of 600 mg divided into two doses, to be given from the day preceding the preparative regimen and continued for the first 3 months of transplantation. UDCA prophylaxis is well tolerated with limited side-effects (46-48).
Heparin and low molecular weight heparin. The potential benefit of prophylactic heparin or low molecular weight heparin in the transplant process is limited given the absence of sufficient studies. Two randomized controlled trials evaluated heparin for hepatic SOS prophylaxis but the results were not consistent (49, 50). It is suggested that patients undergoing autologous HSCT receive low-dose heparin (100 units/kg per day by continuous intravenous infusion) to be started the first day of the preparative regimen and continued until hematopoietic engraftment (51, 52).
Treatment
It is important to note that treatment is tailored according to the severity of the disease. Mild grade disease is usually limited and reversible by supportive management. However, more aggressive disease requires for prompt pharmacological intervention. Although many agents have been tried (including tissue plasminogen activator and steroids), none of these have shown any efficacy.
Defibrotide is the only medication approved by the FDA for the treatment of SOS. Defibrotide is sodium salt of complex single-stranded oligodeoxyribonucleotides derived from porcine mucosal DNA. It is thought that defibrotide has an endothelial protective role and helps the restoration of the thrombotic–fibrinolytic balance, yet the full mechanism of action is poorly defined. It has both anti-inflammatory and anticoagulant activities. It increases the level of tissue plasminogen activator, thus increasing fibrinolysis (53, 54). Defibrotide is included in the guidelines as the only therapeutic pharmacologic agent for SOS but it has not yet been approved as a prophylactic treatment. In March 2016, The FDA approved Defitelio (defibrotide sodium) for the treatment of adult and pediatric patients with hepatic SOS, with renal or pulmonary dysfunction following HSCT (55). The approval was based on the result of three studies evaluating the efficacy of defibrotide sodium in a total of 528 patients. The survival rates on day 100 of therapy were 38% (95% confidence interval=29-48%), 44% (95% confidence interval=33-55%) and 45% (95% confidence interval=40-51%) for each of the studies, respectively (56-58). Based on published reports and analyses of patient-level data, the day 100 survival rates were 21% to 31% for patients with hepatic SOS with renal or pulmonary dysfunction who received supportive care or interventions other than defibrotide sodium. The medication was overall well tolerated with hypotension, diarrhea, vomiting, nausea, and epistaxis being the most common noted adverse reactions. The recommended dose and schedule for defibrotide sodium is 6.25 mg/kg intravenously every 6 hours given as a 2-hour infusion for at least 21 days, and continued until SOS resolution or up to 60 days of treatment (59).
Conclusion
SOS is one of the most dreadful complications of HSCT. Contrary to initial belief, sinusoidal obstruction is not the result of a thrombotic event, but is related to the occlusion of hepatic venules by necrotic cells as a result of an endothelial damage triggered by an inflammatory cascade following toxic injury to hepatic venules. Few preventive measures can be taken, with UDCA to be used in allogeneic HSCT and low-dose heparin in autologous HSCT. Defibrotide is the only medication approved for the treatment of SOS. Physician awareness of predisposing factors and serious attempts to minimize these risks should help reduce the incidence of severe disease.
Footnotes
Conflicts of Interest
The Authors declare that they have no financial relationship with any entity and that this article is a reflection of their own work. This article was written in compliance with all ethical standards.
- Received January 20, 2018.
- Revision received March 4, 2018.
- Accepted March 7, 2018.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved