Durvalumab is an immune checkpoint inhibitor (ICI) that selectively binds to Programmed cell-death 1 ligand 1 (PD-L1) and inhibits its interactions with Programmed cell death-1 (PD-1) and CD80 resulting in T cell activation and anti-tumor responses (1). Durvalumab is currently approved as a single agent for non-small cell lung cancer (NSCLC), in combination with other agents for small cell lung cancer (SCLC) and most recently in 2022, gained approval for advanced biliary tract cancer in combination with chemotherapy (2, 3). Notably, durvalumab is the first immunotherapy to be approved for biliary tract cancer. Durvalumab monotherapy was previously approved for the treatment of advanced urothelial carcinoma but voluntarily withdrawn from the market after failing to meet its primary endpoint in follow-up studies (4).

ICIs are associated with a plethora of immune-related adverse events (irAE) caused by global immune system activation due to the inhibition of the “brakes” – PD-1/PD-L1 or CTLA-4 pathways – of the immune system (5). Immune-mediated liver injury due to ICIs (ILICI) is a subset of drug induced liver injury (DILI) resulting in indirect hepatoxicity presumably via intrahepatic T-cell activation (5, 6). In clinical trials, liver injury is defined based on laboratory cut-offs from the National Cancer Institute’s (NCI) Common Terminology Criteria for Adverse Events (CTCAE) (7). The degree of liver biochemistry and bilirubin elevations determine the grade of liver injury and recommendations of when to withhold or withdraw ICI therapy (8). The American Society of Clinical Oncology (ASCO) recommends evaluation for other causes of liver injury in ICI treated patients that develop elevated liver biochemistries (8). However, the incidence of ILICI compared to other causes of liver injury in patients receiving durvalumab based immunotherapy is not well defined.

Liver injury was rare in the initial clinical trials of durvalumab for NSCLC, SCLC and urothelial carcinoma, occurring in less than 1% of treated patients but elevated liver chemistries were more frequently encountered in patients with advanced biliary tract cancer (4, 9–15). Despite the low overall incidence of liver injury reported in these studies, severe or even fatal hepatic events have been reported in post-marketing surveillance (3). However, there is very limited data on the etiology of the liver injury in these trials (ILICI vs other causes) and outcome of these patients. The aim of our study was to determine the incidence, etiology, and outcomes of liver injury in 112 consecutive patients receiving durvalumab based immunotherapy at a single center who were evaluated and managed using a standardized protocol.

The number of cancer patients eligible for immunotherapy in the United States has dramatically increased from 1.5% in 2011 to an estimated 43.6% of all oncology patients in 2018, with numbers on the rise (18, 19). As of June 2022, durvalumab is currently being studied in 625 clinical trials across the globe (clinicaltrials.gov). Recent studies have shown that clinical features such as patient age are substantially higher in clinical practice compared to clinical trials, and overall survival may be diminished (19, 20). The incidence of hepatotoxicity and ILICI varies widely from 0.7% to 16% depending on the individual agent, its dose, and whether it is given in combination with an anti-CTLA4 agent (21). Therefore, there is an urgent unmet need to better understand the risk factors for developing liver injury and outcomes of patients on durvalumab therapy in clinical practice.

In this study, we describe the liver injury encountered in a diverse group of 112 cancer patients at a single center receiving durvalumab-based treatment, with nearly 50% having lung cancer and the remainder having various other solid organ tumors. Approximately 19% of these patients developed laboratory evidence of liver injury with a 5.4% incidence of DILI from any drug and 3.6% incidence of ILICI specifically attributed to durvalumab as determined by expert opinion and RUCAM scoring.

Interestingly, all cases of ILICI were in non-lung cancer patients in this cohort. A review of the published literature ( Table 4 ) indicates that the overall incidence of liver injury based on the CTCAE grading system in the clinical trials of durvalumab was low, though higher in patients with biliary tract cancers (4, 9–15). The higher incidence of liver injury observed in our patients may be due, in part, to the heterogeneity of our patient population, given the inclusion of patients with HCC and preexisting liver disease who may be at increased risk of developing elevations in their liver chemistries from their underlying cancers. Given the recent approval of durvalumab in combination with chemotherapy for biliary tract cancer, it is important to understand how underlying liver and biliary tract disease may result in abnormal liver enzymes so that these patients do not have durvalumab prematurely discontinued due to presumed ILICI.

In general, the ILICI experienced by our six patients was mild, arising at a median of 35 days after the initial durvalumab infusion, and associated with favorable outcomes. Only 1 of the patients developed jaundice with a peak bilirubin of 2.6 mg/dL, and none required hospitalization or died of hepatotoxicity. Durvalumab was discontinued in two patients due to ILICI based on the recommendations for CTCAE grade 3 hepatoxicity (3, 8). Furthermore, all of the DILI patients experienced normalization of their liver biochemistries during follow-up. These data provide reassurance to clinicians with an overall favorable hepatoxicity profile of durvalumab with generally mild and reversible liver injury, whether used alone or in combination with other agents.

Development of liver injury while on durvalumab immunotherapy was associated with reduced patient survival ( Figure 2 ). Recently, the presence of hepatic metastases have been shown to limit the response to ICI and result in both local and systemic T cell loss (22). Presence of hepatic metastases in comparison to other metastatic sites of disease led to diminished responses to ICI and worsened overall survival (22). Given the high prevalence of hepatic metastases in the liver injury group compared to the non-liver injury group (52.4% vs. 16.5%), we sought to determine if the if the hepatic metastases was driving the worsened survival of this group using multivariate analysis. While the presence of baseline hepatic metastases was indeed predictive of death, liver injury during treatment was independently associated with reduced survival. Further studies are needed to understand how liver injury may independently alter the effectiveness of ICI therapy as we did note increased tumor progression and decreased remission of disease in patients who developed liver injury.

Although baseline liver biochemistries were similar in patients who did and did not develop liver injury, the 6 DILI cases had significantly higher baseline AST and ALT compared to the 15 non-DILI cases ( Table 2 ). Pretreatment laboratory exclusion criteria to receive ICI therapy have ranged from between > 1.5 to > 3x ULN in patients without hepatic metastases and >5x ULN for patients with hepatic metastases (5). Importantly, these patients still responded appropriately to corticosteroids, normalized their liver biochemistries, and did not have worse outcomes, suggesting that baseline liver biochemical abnormalities in these patients are not predictive of worse liver outcomes. In our study cohort, only one patient was rechallenged with durvalumab after treatment with corticosteroids.

Strengths of our study include the significant number of patients who were managed using a common clinical protocol at a single center with a median follow-up of over 1.2 years. However, a large proportion of patients were enrolled in investigational studies with heterogenous tumors and the number of ethnic minority patients was small limiting the generalizability of our findings. Nonetheless, to our knowledge, this is the largest case series describing hepatic outcomes in durvalumab treated patients in clinical practice ( Table 4 ) (23, 24). Following formal causality assessment with expert opinion adjudication and confirmation with RUCAM scores, only six of the 21 liver injury cases (28%) were attributed to DILI and the remainder were attributed to disease progression or other causes ( Figure 3 ). These data are similar to our recent study of pembrolizumab therapy in 420 patients with melanoma and other solid tumors (25). An additional strength of our study includes the use of the well-validated RUCAM scoring system for ICI liver toxicity (26, 27).

An important limitation of our study is the small number of verified DILI cases in this cohort of 112 patients. The retrospective nature of this study also limits our ability to establish causation. The use of EMR mining for case identification may also lead to biases and loss of interpretation of clinical context (28). Additionally, we were limited by the absence of liver biopsies and incomplete serologic evaluations to diagnose alternative etiologies of liver injury. However, recent retrospective studies have indicated that liver biopsy does not necessarily influence the outcome or management of patients with grade 3 or 4 hepatotoxicity and that corticosteroids can be initiated in patients with moderate hepatocellular injury while other causes are being investigated (29). Furthermore, the histology of ILICI has been fairly heterogeneous to date so that liver biopsy may not necessarily impact clinical management. Our data support these recommendations as patients with ILICI were able to be identified by careful review of their laboratory values, serologies, and imaging by their medical oncologist.

In summary, 19% of patients receiving durvalumab-based immunotherapy developed laboratory evidence of liver injury. The liver injury was attributed to DILI in only a small proportion of these cases (26%), highlighting the need for a thorough evaluation of alternative causes of liver biochemistry elevations in oncology patients. All 4 patients with durvalumab ILICI had mild disease and recovered during follow-up. Our data and those from published clinical trials suggest that durvalumab-based immunotherapy has a favorable liver safety profile assuming that vigilant laboratory and clinical monitoring is undertaken.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by University of Michigan Institutional Review Board. Written informed consent from the participants’ legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements.

LS, conception and design, provision of patients, collection and assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript. IK, collection and assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript. IT, data analysis and interpretation, and final approval of manuscript. BS, conception and design, data interpretation, and final approval of manuscript. RF, conception and design, provision of patients, data analysis and interpretation, manuscript writing, and final approval of manuscript. All authors contributed to the article and approved the submitted version.

RF has research supported by Gilead however this is unrelated to the topic of this manuscript.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.