Hypertransaminasemia in cancer patients receiving immunotherapy and immune-based combinations: the MOUSEION-05 study

Abstract

Background

The antitumor efficacy of immune checkpoint inhibitors (ICIs) has increasingly emerged during the last few years. However, there is a need to identify the safety profile of these agents more comprehensively, including liver toxicity.

Materials and methods

Herein, we performed a meta-analysis to assess the risk of all-grade and grade 3–4 hypertransaminasemia in cancer patients receiving ICIs—as monotherapy or in combination with other anticancer agents. All the relevant trials were retrieved through EMBASE, Cochrane Library, and PubMed/Medline databases; eligible studies were selected according to PRISMA statement. The pooled relative risk (RR) and 95% confidence interval (CI) were extracted.

Results

Fifty-nine studies were included. The pooled RRs for all-grade AST and ALT increase were 1.45 (95% CI 1.26–1.67) (Supplementary Fig. 3) and 1.51 (95% CI 1.29–1.77) in patients receiving ICIs monotherapy and immune-based combinations compared to control treatment, respectively. The pooled RRs for grade 3–4 AST and ALT increase were 2.16 (95% CI 1.77–2.64) and 2.3 (95% CI 1.91–2.77).

Conclusions

According to our results, ICIs monotherapy and immune-based combinations were associated with higher risk of all-grade and grade 3–4 hypertransaminasemia. Monitoring liver function should be recommended in cancer patients treated with ICIs monotherapy or immune-based combination, and in case of underlying liver disease, a careful risk–benefit assessment appears as a mandatory need.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00262-023-03366-x.

Introduction

The global cancer burden is destined to increase in the next few years, and cancer represents a serious threat to public health worldwide. In fact, the rising proportion of adults and elderly patients are two important factors underlying the increase in cancer-related mortality, and although the incidence of some solid tumors has decreased in some countries, the limited efficacy of systemic treatments in several setting and diseases remains a crucial issue [1].

The advent of immune checkpoint inhibitors (ICIs) has represented a therapeutic revolution over the last ten years [2], with immunotherapy representing an effective treatment strategy in multiple hematological and solid tumors, as reported in several practice-changing clinical trials [3–7]. ICIs, whose mechanism of action is based on the activation of the immune system, are able to modulate T lymphocytes and to target immune checkpoints such as programmed cell death protein 1 (PD-1), programmed cell death ligand 1 (PD-L1), and cytotoxic t lymphocyte-associated protein 4 (CTLA-4). ICIs, as monotherapy or in combination with other anticancer agents, have achieved unprecedented results in melanoma, renal cell carcinoma (RCC), non-small cell lung cancer (NSCLC), hepatocellular carcinoma (HCC), colorectal cancer, urothelial carcinoma (UC), and several others [8–14]. Although the safety profile of ICIs monotherapy and immune-based combinations is considered acceptable, ICIs have a specific set of treatment-related adverse events, which are commonly defined as immune-related adverse events (irAEs). IrAEs are determined by an erroneous activation of the immune system and may affect multiple organ systems, ranging from liver to thyroid, lung, pancreas, and skin [15, 16]. In addition, the type and grade of irAEs may widely vary according to the type of ICI, the tumor type, and the immune-based combination, since the concomitant use of systemic chemotherapy or tyrosine kinase inhibitors may increase not only the anticancer activity but also adverse events, which sometimes are even life-threatening [17–19].

According to findings from clinical trials, ICIs are deemed to determine hepatotoxicity in a percentage ranging from 1 to 10% of cases, with patients often presenting with elevation of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) [20]. From a temporal perspective, these alterations are frequently observed between 6 and 14 weeks after the start of treatment and most of these adverse events are managed by corticosteroid use, while the addition of other immunosuppressive treatments is evaluated in case of lack of response to corticosteroids [21, 22]. Since the use of ICIs and immune-based combinations is destined to grow in cancer patients, further investigations to comprehensively characterize the impact of these events are needed. Based on these premises, in the MOUSEION-05 we aimed at investigating the risk of all-grade and grade 3–4 hypertransaminasemia in randomized controlled clinical trials (RCTs) assessing ICIs monotherapy and immune-based combinations in cancer patients.

Materials and methods

Search strategy

All phase II and III clinical trials published from June 15, 2008, to November 10, 2022, evaluating immunotherapy and immune-based combinations in cancer patients were retrieved by five different authors. Keywords used for searching on EMBASE, Cochrane Library, and PubMed/ Medline were the following: “immunotherapy” OR “nivolumab” OR “ipilimumab” OR “atezolizumab” OR “pembrolizumab” OR “durvalumab” OR “avelumab” OR “tremelimumab” OR “immune checkpoint inhibitors” OR “immunotherapy” AND “hypertransaminasemia” OR “increased AST” OR “increased ALT” OR “hepatotoxicity” AND “cancer” OR “cancer patients.” Only articles written in the English language and published in peer-reviewed journals were included. Proceedings of the main international oncological meetings (such as European Society of Medical Oncology [ESMO], American Society of Clinical Oncology [ASCO], American Association for Cancer Research [AACR], and European CanCer Organization [ECCO]) were also searched from 2008 onward for relevant trials and/or abstracts.

Selection criteria

Trials retrieved from the first analysis we conducted were subsequently restricted to: 1) prospective phase II and III RCTs in cancer patients; 2) participants receiving ICIs monotherapy or immune-based combinations; and 3) studies with available data in terms of all-grade and grade 3–4 hypertransaminasemia in the immune-based combination group and the control arm. Clinical trials with two different immunotherapy arms were split in two sections in the analysis.

Data extraction

The following data were extracted for each publication: 1) RCT general information (author, year, phase); 2) interventions and dosage of drugs; 3) number of cancer patients; and 4) available outcomes in terms of all-grade and grade 3–4 hypertransaminasemia.

Available outcomes were measured as relative risks (RRs) and 95% confidence intervals (CIs). Five separate authors conducted the search and identification independently. The meta-analysis was conducted according to Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines (Supplementary) [23].

Statistical design

All statistical analyses were performed using R Studio.

RRs were used to analyze dichotomous variables, including all-grade and grade 3–4 hypertransaminasemia in cancer patients treated with ICIs monotherapy and immune-based combinations versus control treatment. Forest plots were used to assess RRs. Statistical heterogeneity between the included trials was examined using the Chi-square test and the I² statistic; substantial heterogeneity was considered to be present when the I² value was greater than 50% or there was a low P value (< 0.10) in the Chi-square test [24]. When no heterogeneity was noted, the fixed effects model was used, while the random-effects model was applied in the presence of significant heterogeneity.

Results

Selected studies

A total of 6586 potentially relevant reports were identified, which were restricted to 59 following independent evaluation of five authors [25–82]. We excluded 6522 records as non-pertinent reports (editorials, case reports, ongoing studies/trials in progress, review articles, preclinical studies, retrospective studies, systematic reviews and meta-analyses, single-arm trials, non-randomized trials, etc.). Eligible studies were identified and selected as shown in Supplementary Fig. 2; a summary of the included trials is presented in Table 1.

Table 1.

Summary of the included studies

Reference	Phase	Malignancy	Treatment arm	Control arm
Powles T et al. (A) [25]	3	UC	Durvalumab + tremelimumab	Chemotherapy
Powles T et al. (B) [25]	3	UC	Durvalumab	Chemotherapy
Goldman et al. (A) [26]	3	SCLC	Durvalumab + carboplatin + etoposide	Carboplatin + etoposide
Goldman et al. (B) [26]	3	SCLC	Durvalumab + tremelimumab + carboplatin + etoposide	Carboplatin + etoposide
Renouf DJ et al [27]	2	PDAC	Durvalumab + tremelimumab + chemotherapy	Chemotherapy
Robert C et al [28]	3	Melanoma	Nivolumab	Dacarbazine
Borghaei H et al [29]	3	Non-squamous NSCLC	Nivolumab	Docetaxel
Brahmer J et al [30]	3	Squamous NSCLC	Nivolumab	Docetaxel
Larkin J et al [31]	3	Melanoma	Nivolumab	Chemotherapy
Gillison ML [32]	3	HNT	Nivolumab	Chemotherapy
Cella D et al [33]	3	RCC	Nivolumab + ipilimumab	Sunitinib
Wu YL et al [34]	3	NSCLC	Nivolumab	Docetaxel
Reardon DA et al [35]	3	Glioblastoma	Nivolumab	Bevacizumab
Janjigian YY et al [36]	3	GC/GEJC	Nivolumab + chemotherapy	Chemotherapy
Choueiri T et al [37]	3	RCC	Nivolumab + cabozantinib	Sunitinib
Hamanishi et al [38]	3	Ovarian cancer	Nivolumab	Chemotherapy
Hayashi H et al [39]	3	NSCLC	Nivolumab	Chemotherapy
Schmid P et al [40]	3	TNBC	Atezolizumab + nab-paclitaxel	Nab-paclitaxel
Socinski MA et al [41]	3	Non-squamous NSCLC	Atezolizumab + bevacizumab + chemotherapy	Bevacizumab + chemotherapy
Powles T et al [42]	3	UC	Atezolizumab	Chemotherapy
Eng C et al. (A) [43]	3	CRC	Atezolizumab + cobimetinib	Regorafenib
Eng C et al. (B) [43]	3	CRC	Atezolizumab	Regorafenib
West H et al [44]	3	Non-squamous NSCLC	Atezolizumab plus chemotherapy	Chemotherapy
Finn RS et al [45]	3	HCC	Atezolizumab + bevacizumab	Sorafenib
Mittendorf EA et al [46]	3	TNBC	Atezolizumab + chemotherapy	Chemotherapy
Herbst RS et al [47]	3	NSCLC	Atezolizumab	Chemotherapy
Emens LA et al [41]	2	HER2 + BC	Atezolizumab + TDM1	Placebo + TDM1
Felip E et al [48]	3	NSCLC	Atezolizumab	BSC
Bellmunt J et al [49]	3	UC	Atezolizumab	Observation
Miles D et al [50]	3	TNBC	Atezolizumab + chemotherapy	Placebo + chemotherapy
Cho BC et al [51]	2	NSCLC	Atezolizumab + tiragolumab	Atezolizumab + placebo
Kelley RK et al [52]	3	HCC	Atezolizumab + cabozantinib	Sorafenib
Mettu NB et al [53]	2	CRC	Atezolizumab + bevacizumab + chemotherapy	Chemotherapy + bevacizumab
Huober J et al [54]	3	HER2 + BC	Atezolizumab + pertuzumab + trastuzumab + chemotherapy	Placebo + pertuzumab + trastuzumab + chemotherapy
Pal SK et al [55]	3	RCC	Atezolizumab	Placebo
Powles T et al [83]	3	CRPC	Atezolizumab + enzalutamide	Enzalutamide
Bang YJ et al [56]	3	GC/GEJC	Avelumab	Chemotherapy
Barlesi F et al [57]	3	NSCLC	Avelumab	Docetaxel
Motzer RJ et al [58]	3	RCC	Avelumab + axitinib	Sunitinib
Powles T et al [59]	3	UC	Avelumab	BSC
Moehler et al [60]	3	GC/GEJC	Avelumab	Chemotherapy
Lee NY et al [61]	3	HNT	Avelumab + chemoradiotherapy	Chemoradiotherapy
Pujade-Lauraine E. et al [62]	3	Ovarian cancer	Avelumab	PLD
Pujade-Lauraine E. et al [62]	3	Ovarian cancer	Avelumab ± PLD	PLD
Monk BJ et al [63]	3	Ovarian cancer	Chemotherapy—> manteinance avelumab	Chemotherapy—> observation
Monk BJ et al [63]	3	Ovarian cancer	Chemotherapy + avelumab—> manteinance avelumab	Chemotherapy—> observation
Redman J et al [64]	2	CRC	mFOLFOX + bevacizumab + AdCEA vaccine + avelumab	mFOLFOX + bevacizumab
RS Herbst et al [65]	2/3	NSCLC	Pembrolizumab	Docetaxel
Mok TSK et al [66]	3	NSCLC	Pembrolizumab	Chemotherapy
Bellmunt J et al [67]	3	UC	Pembrolizumab	Chemotherapy
Powles T et al [68]	3	RCC	Pembrolizumab + axitinib	Sunitinib
Schmid P et al [69]	3	TNBC	Pembrolizumab + chemotherapy	Chemotherapy
Ferrucci PF et al [70]	2	Melanoma	Pembrolizumab + dabrafenib + trametinib	Placebo + dabrafenib + trametinib
Tolaney SM et al [71]	2	Breast Cancer	Pembrolizumab + eribulina	Eribulina
Motzer RJ et al [72]	3	RCC	Pembrolizumab + lenvatinib	Everolimus + lenvatinib / sunitinib
Awad MM et al [73]	2	Non-squamous NSCLC	Pembrolizumab + chemotherapy	Chemotherapy
Colombo N et al [74]	3	Cervical Cancer	Pembrolizumab + chemotharapy + bevacizumab	Placebo + chemotherapy + bevacizumab
Powles T et al. (A) [75]	3	UC	Pembrolizumab + chemotherapy	Chemotherapy
Powles T et al. (B) [76]	3	UC	Pembrolizumab	Chemotherapy
Sun JM et al [77]	3	Oesophageal Cancer	Pembrolizumab + chemotherapy	Placebo + chemotherapy
Winer EP et al [76]	3	TNBC	Pembrolizumab	Chemotherapy
Diaz LA et al [78]	3	CRC	Pembrolizumab	Chemotherapy
Makker V et al [79]	3	Endometrial cancer	Pembrolizumab + lenvatinib	Chemotherapy
Chung HC et al [80]	3	GC/GEJC	Pembrolizumab	Chemotherapy
Reckamp KL et al [81]	2	NSCLC	Pembrolizumab + ramucirumab	Chemotherapy

BC breast cancer, CRC colorectal cancer, GC/GEJC gastric cancer/gastroesophageal cancer, HCC hepatocellular carcinoma, HNT head and neck cancer, NSCLC non-small cell lung cancer, PDAC pancreatic ductal adenocarcinoma, RCC renal cell carcinoma, SCLC small cell lung cancer, TNBC triple negative breast cancer, UC urothelial cancer

All-grade AST and ALT increase

The pooled RRs for all-grade AST and ALT increase were 1.45 (95% CI 1.26–1.67) (Supplementary Fig. 2) and 1.51 (95% CI 1.29–1.77) (Supplementary Fig. 3), in patients receiving ICIs monotherapy and immune-based combinations compared to control arm, respectively. Both analyses were associated with substantial heterogeneity (I² of 57% and 64%, respectively), and thus, a random effect model was used.

Grade 3–4 AST and ALT increase

The pooled RRs for grade 3–4 AST and ALT increase were 2.16 (95% CI 1.77–2.64) (Fig. 1) and 2.3 (95% CI 1.91–2.77) (Fig. 2) for patients treated with ICIs monotherapy and immune-based combinations compared to control arm, respectively. A fixed effect model was used since the analyses presented low heterogeneity.

Fig. 1.

Forest plot of comparison between ICIs monotherapy and immune-based combinations versus control arm in cancer patients; the outcome was the relative risk (RR) of grade 3–4 AST increase

Fig. 2.

Forest plot of comparison between ICIs monotherapy and immune-based combinations versus control arm in cancer patients; the outcome was the relative risk (RR) grade 3–4 ALT increase

Discussion

The impact of ICIs monotherapy and immune-based combinations has represented an unprecedented step forward in the systemic treatment for cancer patients, with these agents which have been associated with robust, and even long-lasting responses [82–84]. However, irAEs may represent an issue in this setting, and few data are available in terms of risk and incidence of hepatotoxicity, including increased low-grade and high-grade AST and ALT serum levels [85, 86]. To the best of the authors’ knowledge, our study represents the most comprehensive and up-to-date meta-analysis in the literature to systematically assess the risk of all-grade and grade 3–4 hypertransaminasemia in cancer patients receiving ICIs monotherapy and immune-based combinations. In MOUSEION-05, more than 30.000 patients from 59 clinical trials were included in the analysis. Higher risk of all-grade and grade 3–4 hypertransaminasemia was reported in cancer patients treated with immunotherapy compared with control treatments. Of note, the highest RR was observed for grade 3–4 AST and ALT increase–2.16 (95% CI 1.77–2.64) and 2.3 (95% CI 1.91–2.77), respectively.

Among irAEs, liver toxicity is frequently underestimated due to several reasons, including its ambiguous clinical presentation and its lower incidence compared with other common adverse events [87]. In fact, all-grade and grade 3–4 hypertransaminasemia are commonly unnoticed since these events may be totally asymptomatic or paucisymptomatic. Given the increased incidence of these toxicities, monitoring liver function should be recommended in cancer patients treated with ICIs monotherapy or immune-based combination, and in case of underlying liver disease, a careful risk–benefit assessment appears as a mandatory need. In addition, clinicians are called to suspect and to be aware of these events associated with immunotherapy, and liver function test, physical examination, clinical judgment, and medical history—before the start as well as during immune-based combinations—are fundamental tools in cancer immunotherapy. ICIs monotherapy and immune-based combinations are now the standard for the treatment of several tumor types, as recommended by several international guidelines following the results of landmark clinical trials in this setting [88–92]. However, these treatments have not yet been approved in several countries or have been approved in the last few months, and thus, clinicians have still limited experience of immune-based combinations in everyday clinical practice. Thus, clinical experience plays and will play a fundamental role in the balance between antitumor activity, safety, and quality of life. At the same time, the timing itself of these irAEs is important to determine whether these toxicities are due to ICIs or other anticancer drugs in patients treated with combination strategies, something that changes the management itself of toxicities.

Some strengths and limitations of our meta-analysis should be highlighted. Among the strengths of this study, our analysis included 59 phase II and III RCTs by using the most updated data in terms of all-grade and grade 3–4 hypertransaminasemia. In addition, we included an overall large number of cancer patients. However, some limitations should also be underlined. First, the current meta-analysis was based on pooled data, and thus, the presence of single-patient variables was not included. Secondly, although the random-effects model was performed to reduce heterogeneities across studies, the analysis regarding all-grade AST and ALT increase presented substantial heterogeneity, and thus, our findings should be interpreted cautiously. Thirdly, the trials included in the meta-analysis evaluated widely different ICIs monotherapy and immune-based combinations. All these anticancer agents—ranging from systemic chemotherapy to tyrosine kinase inhibitors and other antiangiogenic drugs—present different and not superimposable safety profiles, and thus, this element could have produced some bias affecting our results. Moreover, a possible overlap in side effects between different agents should be considered, something that could have introduced some bias. For example, it is not possible to clarify with absolute certainty if a specific event is related to immunotherapy or other anticancer drugs included in immune-based combinations as well as other concomitant medications and dietary and lifestyle factors. Similarly, the data were extracted from registration trials, where the time of the onset of the event was not reported. In addition, the presence of selection bias cannot be excluded, since all cancer patients were selected subjects enrolled into high-quality trials conducted at academic centers and with good performance status. Publication bias due to the higher publishing possibility of studies with positive results could not be excluded. Lastly, our study was not a network meta-analysis, and thus, it was not possible to determine which combination may be more hepatotoxic than the other.

However, despite the limitations affecting the current analysis, we think that this piece of evidence emphasizes both the importance of a correct patient's stratification due to any underlying liver impairment, as well as the need to carefully supervise regular blood tests and to better customize therapeutic decisions. Beyond the efficacy of these novel combination strategies, clinicians should pay attention to these frequently underestimated adverse events.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

A.R. wrote the main manuscript text and performed the statistical analysis. A.R. prepared figures 1-5 and table 1. V.M., V.T., E.T., A.M., and M.R. contributed to data acquisition. V.M., R.D.L., M.S., and F.M. contributed to the conception, design, and manuscript revision. All authors approved the final version of the manuscript.

Declarations

Conflict of interest

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript.