Management of Hepatitis B Virus and Hepatitis C Virus Infections in Patients with Cancer Receiving Immune Checkpoint Inhibitors Open Access

In the United States, 850,000 to 2.2 million people are chronically infected with hepatitis B virus (HBV), and 2.4 million are chronically infected with hepatitis C virus (HCV).^[1–4] Among newly diagnosed patients with cancer in the United States, 6.5% have past HBV infection, 0.6% have chronic HBV infection, and 2.4% have HCV infection.^[5]

Immune checkpoint inhibitors (ICIs), such as programmed death receptor-1 (PD-1) inhibitors, programmed death receptor ligand-1 (PD-L1) inhibitors, and cytotoxic T-lymphocytic antigen 4 (CTLA-4) inhibitors, are part of standard-of-care treatment for many solid tumors and hematologic malignancies.^[6] Many cancer clinical trials of ICIs exclude patients with HBV infection or HCV infection because of safety concerns regarding hepatitis flare and viral reactivation.^[7]

The safety of ICI therapy in patients with cancer and chronic HBV or HCV infection is poorly defined. There is a risk of liver injury grade 3 and higher in patients with chronic HBV or HCV infection during ICI therapy.^[8–10] Chronic HBV or HCV infection leads to virus-specific CD8+ T-cell exhaustion, decreases cytokine (interferon-gamma) secretion, and increases immune checkpoint expression, and could lead to liver injury in patients with cancer receiving ICIs.^[11–13] In addition, immunosuppression used for managing immune-related toxic effects following ICI therapy may cause HCV reactivation, although it remains unclear whether immunosuppression may also cause HBV reactivation.^[12,14,15] Patients with chronic HBV infection who are receiving ICIs may experience a flare-up of HBV replication, which can cause severe liver damage, liver failure, or even death.^[16,17] Despite the unclear safety of ICIs in patients with HBV or HCV infection, recently published data from a European survey of 56 oncology centers showed that only one-third of the respondents would start antiviral prophylaxis for chronic HBV at ICI initiation.^[18]

Given that patients with HBV and HCV infection are excluded from many cancer clinical trials, there is a paucity of data on HBV and HCV reactivation in patients with cancer treated with ICIs.^[12,19,20] To address this gap, we reviewed the literature on the use of ICIs among patients with HBV or HCV infection and provide general recommendations on management of HBV and HCV infections in patients with cancer receiving ICIs.

We searched MEDLINE and PubMed for all original research articles, case reports, and systematic reviews published in English between July 1, 2013, and July 7, 2023, on patients with cancer and HBV or HCV infection receiving ICIs. In this article, we follow the standard definitions of HBV reactivation recommended by the American Association for the Study of Liver Diseases and the American Society of Clinical Oncology (Table 1).^[16,21] Chronic HBV infection is defined as positivity for hepatitis B surface antigen (HBsAg) for more than 6 months. Past HBV infection is defined as negativity for HBsAg and positivity for hepatitis B core antigen (anti-HBc), irrespective of the status of antibodies against HBsAg.^[16,21]

The reactivation rates were analyzed by using descriptive statistics. All descriptive statistical analyses were performed with STATA IC software, version 12.0 (StataCorp LP, College Station, TX). Descriptive p-values at a two-sided significance level of 0.05 were reported.

Among patients with solid tumors, the risk of HBV reactivation due to ICI therapy is 4% in patients with chronic HBV infection and 0% in patients with past HBV infection.^[19,22] HBV reactivation after ICIs can be fatal, but recent studies have not resulted in any cases of HBV-related liver failure or death.^[7] In our search of the literature, we found 28 studies that evaluated the safety of ICI therapy in patients with HBV infection and cancer: 3 prospective clinical trials, 7 retrospective cohort studies, 9 retrospective case series, and 9 case reports (Table 2).^{[7,8,20,23–45]} We also found two systematic reviews on this topic.^[15,19]

In a systematic review of 633 patients with advanced cancer and HBV infection receiving ICIs, 26 patients (4.1%), all with chronic HBV infection, developed HBV reactivation, and no HBV-related deaths were reported. No cases of reactivation occurred in patients with past HBV infection.^[19]

In a systematic review and meta-analysis of 1057 patients with advanced cancer and chronic HBV infection receiving ICIs, 18 patients (1.7%) developed HBV reactivation. The risk of HBV reactivation was high (11%) in patients with chronic HBV infection not receiving antiviral prophylaxis. The authors stated that among patients who receive only ICIs, immune-related toxic effects that require high-dose corticosteroids might reduce the efficacy of ICIs, increasing HBV reactivation risk. No HBV-related deaths were reported.^[15]

The three prospective clinical trials^[23,24,45] specifically targeted patients with advanced hepatocellular carcinoma (HCC) and enrolled patients with HBV or HCV infection. These patients were required to receive antiviral treatment and maintain a viral load of less than 100 IU/mL.^[23,24,45] There were no cases of HBV-related liver failure or death. No grade 3–4 liver-related adverse events were seen.

CheckMate 040^[23] was an open-label, noncomparative, phase 1/2 trial in which 262 patients with HCC were treated with nivolumab. Sixty-six patients had chronic HBV infection, but no cases of HBV reactivation were observed.

KEYNOTE-224^[24] was a nonrandomized, open-label, phase 2 trial in which 104 patients with advanced HCC were treated with pembrolizumab. Twenty-two patients had chronic HBV infection, but no cases of HBV reactivation or hepatitis flare occurred.

CheckMate 459^[45] was a randomized, multicenter, open-label, phase 3 trial in which 371 patients with HCC were treated with nivolumab. A total of 116 patients had chronic HBV infection, but no cases of HBV reactivation were reported.

In a retrospective study from MD Anderson Cancer Center with 10 patients with chronic HBV infection, no instances of HBV reactivation or hepatitis flare were observed, and no HBV-related deaths were reported. In that study, all patients received antiviral treatment.^[7]

A cohort study with 990 patients (397 with chronic HBV infection and 225 with past HBV infection) demonstrated a low incidence of HBV reactivation during ICI therapy. Specifically, only two patients (1%) with chronic HBV infection and none with past HBV infection experienced reactivation. In that study, all patients with chronic HBV infection were receiving antiviral treatment at the time of HBV reactivation (started before ICI therapy in 388 patients and during ICI therapy in 9 patients).^[8]

In a retrospective cohort study of 60 patients with HCC and chronic HBV infection treated with nivolumab and pembrolizumab, HBV reactivation occurred in only 1 of the 6 patients not receiving antiviral treatment. Among the 54 patients who received entecavir, tenofovir disoproxil fumarate (TDF), or tenofovir alafenamide during ICI therapy, none experienced HBV reactivation.^[26]

In a historical cohort study of 3465 patients with cancer receiving ICIs, including 511 with chronic HBV infection and 564 with past HBV infection, HBV reactivation was rare. Only five patients (1%) with chronic HBV infection and none of those with past HBV infection experienced reactivation. Among the five patients with HBV reactivation, three were not receiving antiviral prophylaxis, one was noncompliant with antivirals, and one had interrupted antiviral prophylaxis. This study demonstrated that HBV reactivation can happen in patients with chronic HBV infection who are noncompliant with antiviral prophylaxis during ICI therapy.^[27]

In a retrospective cohort study of 114 patients with cancer and chronic HBV infection receiving anti–PD-1 or PD-L1 antibody, HBV reactivation occurred in six patients (5.3%). The absence of antiviral prophylaxis was the only significant risk factor for reactivation.^[25] Among the six patients who experienced HBV reactivation, one had received prophylactic entecavir therapy prior to ICI therapy, whereas the remaining five were not receiving prophylaxis. When HBV reactivation was detected, five of the five patients without prophylaxis were started on entecavir. All six patients achieved undetectable HBV DNA levels, and no HBV-related deaths were reported.

In a retrospective cohort study of 42 patients with cancer (8 with chronic HBV infection and 34 with past HBV infection), reactivation was observed in 2 patients (5%) with chronic HBV infection. Among them, one patient without antiviral prophylaxis developed grade 3 liver injury and HBV reactivation, reaching a peak HBV DNA level of 7.09 log IU/mL 1 week after starting pembrolizumab, leading to discontinuation of ICI. This patient was started on steroids and TDF but died of cancer progression. Another patient developed HBV DNA increase despite telbivudine; HBV DNA was closely monitored and became undetectable. No one with past HBV infection experienced reactivation.^[46]

In a retrospective cohort study of 62 patients (7 with chronic HBV infection and 55 with past HBV infection) with HCC treated with ICIs, 6 patients (9.7%) experienced reactivation. Reactivation occurred in five patients with chronic HBV infection and one with past HBV infection. In that study, 92% of patients were receiving antiviral treatment before ICIs, none developed hepatitis failure, and the use of antiviral treatment was associated with a lower risk of liver injury grade 3 and higher (p = 0.048).^[47]

Among the nine retrospective case series (Table 2), there were 10 cases of HBV reactivation after ICI therapy, all in patients with chronic HBV infection.^[20,28–35] In three of these 10 cases, HBV reactivation led to the discontinuation of ICIs. In two of the cases, ICI therapy was discontinued owing to grade 4 hepatitis and grade 2 pneumonitis, respectively; in the third case, ICI therapy was discontinued owing to grade 4 hepatitis while receiving PD-1 inhibitor in combination with hepatic arterial infusion chemotherapy. In all 10 cases, either the patient had received antiviral prophylaxis and it was continued after viral reactivation, or antivirals were started at the time of viral reactivation. Eight of the 10 patients with HBV reactivation experienced an increase in HBV DNA level from undetectable at baseline to greater than 3 log (1000) IU/L after 1 to 4 months of ICI therapy (pembrolizumab in 3, toripalimab in 3, and sintilimab and nivolumab in 1 each) and a decrease in HBV DNA level after initiation of treatment with either entecavir (in 6 patients) or tenofovir disoproxil fumarate (in two patients). One patient with HBV reactivation experienced an increase in HBV DNA level of 2 log (100) IU/mL over baseline after one cycle of nivolumab. One patient with HBV reactivation experienced an increase in HBV DNA level of more than 4 log (10,000) IU/mL after 1 month of pembrolizumab while taking entecavir, but HBV DNA decreased spontaneously without modification of pembrolizumab and entecavir treatment.

The nine case reports^[36–44] regarding the safety of HBV infection after ICI therapy are summarized in Table 1. All 5 patients who did not receive antiviral prophylaxis experienced HBV reactivation, and antivirals were started at the onset of viral reactivation.^[37–40,44] Two of these five patients had chronic HBV infection before ICI therapy targeting anti-PD1 (pembrolizumab) and anti–CTLA-4 + anti-PD1 (ipilimumab plus nivolumab), respectively, and recovered after antiviral treatment with TDF.^[37,38] One patient had past HBV infection before ICI therapy targeting anti-PD1 (nivolumab) and recovered after TDF treatment.^[40] One patient had unknown HBV status before ICI therapy targeting anti–PD-L1 (durvalumab) and experienced severe HBV reactivation with peak alanine aminotransferase level of 1168 IU/L and peak HBV DNA level of 7.01 log IU/mL after initiation of durvalumab.^[44] This patient had a decrease in HBV DNA level after starting TDF but died of sepsis. Four patients had chronic HBV infection before ICI therapy and received antiviral prophylaxis; none of these four patients experienced HBV reactivation or hepatitis flare.^[36,41–43]

These cases provide evidence that HBV reactivation can occur in patients with chronic HBV infection after many different types of ICI therapy. Our findings agree with those of a recent meta-analysis,^[15] and we agree with the authors of that meta-analysis and with a recent American Society of Clinical Oncology (ASCO) provisional clinical opinion update^[21] that concomitant use of antiviral treatment with ICI therapy is safe and necessary in patients with chronic HBV infection.

As of Jul 7, 2023, a total of 2799 patients with cancer with chronic or past HBV infection receiving ICIs had been reported in the literature.^{[7,8,20,23–35,37–40,44–47]} Of these, 1.4% (38/2799) had HBV reactivation, of whom 35 (92%) had chronic HBV infection (Fig. 1). The frequency of HBV reactivation in patients with chronic and past HBV infections was 2% (35/1667) and 0.3% (3/1132), respectively. The risk of HBV reactivation was significantly higher among patients with chronic HBV infection not receiving antiviral prophylaxis than among those receiving antivirals (17% vs 1%, p < 0.05). The ICIs most often associated with reactivation were PD-1 inhibitors, which were implicated in 33 cases (87%) (nivolumab in 15 [39%], pembrolizumab in 11 [29%], toripalimab in 4 [11%], camrelizumab in 2 [5%], and sintilimab in 1 [3%]).

Because of its effectiveness in preventing HBV reactivation, HBV prophylaxis is currently recommended by several professional societies in the United States and Europe^{[15,16,21,48]} for patients with cancer with chronic HBV infection receiving ICIs (Table 3). However, for patients with cancer and past HBV infection receiving ICIs, routine antiviral prophylaxis is not supported by published data. Therefore, it is crucial to screen patients with cancer for HBV infection (through screening for HBsAg, anti-HBc, and hepatitis B surface antibody [anti-HBs]) before starting ICI therapy and to monitor HBV markers, such as HBsAg and HBV DNA, during ICI therapy.^[21]

Theoretically, ICIs have the potential to restore virus-specific T-cell responses and inhibit HBV or HCV replication.^[13,49] HBV inhibition was reported in a cohort of 60 patients with HBV-related HCC receiving ICIs. In that study, three of the six HBV-infected patients not receiving antivirals had a decrease in HBV viral load of more than 1 log during the course of ICI therapy, and one patient consistently showed undetectable levels of serum HBV DNA.^[26] Clinical trials are underway in patients without cancer to try to achieve HBV cure by inhibiting viral replication, reducing viral antigen load, enhancing immune responses, and targeting specific antiviral immune responses.^[50]

Therapeutic vaccines function by inducing CD4 and CD8 T-cell responses and leading to viral clearance. Although therapeutic vaccines and engineered HBV-specific and HCV-specific T cells have reduced HBV and HCV viral loads in preclinical models, results in humans have not yet been confirmed because T-cell defects may be irreversible and may require a long time to recover.^[51,52] Antiviral agents or ICIs alone are insufficient to control T-cell exhaustion, but use of ICIs in combination with new therapeutic vaccines and HBV-specific mediators such as bispecific soluble T-cell receptors or Fc-engineered monoclonal antibodies is being explored.^[53] Currently, human studies are underway to evaluate the efficacy of ICIs in combination with therapeutic vaccines, especially in patients with chronic HBV infection, in an effort to achieve HBV cure.^[49,50,54] Although fewer trials have achieved sustained HBsAg loss, the combination of therapeutic vaccines with antivirals with a high genetic barrier to resistance is also a promising therapeutic approach to achieving a complete cure.^[50,53]

We found 11 studies that evaluated the safety of ICI therapy in patients with HCV infection and cancer: 5 clinical trials, 5 retrospective studies, and 1 prospective observational study (Table 4). We also found a review article on this topic.^[12] The clinical course of HCV reactivation is less severe than that of HBV reactivation.^[55,56] A previous review of seven studies, including 82 patients with cancer with chronic HCV infection treated with ICIs,^[12] showed that ICIs were safe, and viral reactivation did not occur. A multicenter clinical trial and two multicenter retrospective analyses of patients with solid tumors and chronic HCV infection receiving ICIs revealed no cases of viral reactivation or HCV-related death.^[31,57] A prospective study from MD Anderson Cancer Center indicated that ICIs are safe in HCV-infected patients with solid tumors.^[14] In that study, only 2 of 52 patients (4%) treated with ICIs had HCV reactivation, and those patients were also receiving immunosuppressants such as infliximab and high-dose corticosteroids for ICI-related toxic effects, making it difficult to determine if the reactivation was associated with ICI therapy alone.^[58] None of the other studies we found on ICI therapy in HCV-infected patients with cancer showed any cases of HCV reactivation (Table 4).

Oncologic professional societies recommend universal HCV screening for all patients with cancer.^[59,60] Published data^{[6,7,14,23,24,31,32,45,57,61,62]} support that HCV-infected patients with cancer can be safely enrolled in cancer clinical trials of ICIs; however, these patients should be monitored for hepatitis flare and reactivation (or enhanced HCV replication). The management of HCV in patients with cancer receiving ICIs should be individualized on the basis of the patient’s viral hepatitis status, liver function, cancer treatment plan, and cancer status.^[63] Antiviral treatment before, during, or after ICI therapy is safe and can eradicate HCV.^[14] In a recent editorial, it was emphasized that ICI therapy is associated with a favorable prognosis in HCV-infected patients. The authors advocate using ICIs in HCV-infected patients in both clinical practice and cancer clinical trials.^[58] Management of HCV infection in patients with cancer receiving ICIs is shown in Figure 2.

Immune checkpoint blockade in cancer therapy may affect HCV infection, potentially inhibiting HCV replication by reversing T-cell exhaustion, inducing T-cell responses, and promoting viral clearance.^[13,61] HCV inhibition has been reported in 7.7 to 12% of HCV-infected patients receiving ICIs without antivirals.^[14,61]

Recent studies show that HCV-specific CD8+ T cells preserve a molecular hallmark of T-cell exhaustion that may be irreversible even 1 year after HCV cure and may require a long time to recover from exhaustion.^[11,64] However, in the MD Anderson study mentioned above, the authors reported the first case of a patient with cancer and chronic HCV infection who experienced sustained HCV clearance after ICI therapy in the absence of antiviral treatment. In that same study, 4 of 52 patients (8%) treated with ICIs had HCV inhibition in the absence of antivirals.^[14]

ICI therapy seems to be safe and effective in patients with solid tumors with HBV or HCV infection. Treatment with ICIs has been associated with HBV reactivation, but mainly in patients with chronic HBV infection. HBV reactivation can be managed by either closely monitoring with early initiation of antiviral treatment in patients with past HBV infection or antiviral prophylaxis in those with chronic HBV infection. ICIs alone without direct-acting antivirals can inhibit HCV replication and may achieve virologic cure. Treatment with ICIs has been associated with HCV reactivation, but mostly in patients receiving immunosuppressants for ICI-related toxic effects. Chronic HBV or HCV infection should not be considered a contraindication for ICI therapy. More data are needed on the occurrence of HCV inhibition and HCV reactivation in patients receiving ICIs. Knowledge gaps related to HBV and HCV reactivation in patients receiving ICIs are listed in Table 5.

The authors thank Stephanie Deming of the Research Medical Library at MD Anderson for editorial assistance.