A case of toxic epidermal necrolysis associated with lenvatinib and sintilimab therapy for intrahepatic cholangiocarcinoma

Abstract

Lenvatinib is used as the first-line treatment for intrahepatic cholangiocarcinoma. Sintilimab is a programmed cell death receptor-1 (PD-1) antibody used in the treatment of solid tumors. We present the case of a 78-year-old man with fatal toxic epidermal necrolysis (TEN) associated with the use of sintilimab followed by lenvatinib. This patient, who presented with intrahepatic cholangiocarcinoma, first received immunotherapy with sintilimab according to the standard schedule of 200 mg every 3 weeks. The patient began to receive 8 mg of lenvatinib daily 1 day after sintilimab therapy was initiated. Multiple erythematous papules and blisters appeared on the patient’s face and trunk and gradually spread to his arms and legs, and the lesions extensively involved >30% of the body surface area 18 days after lenvatinib initiation. The patient stopped taking lenvatinib on the next day. The skin rash quickly progressed over 1 week to a tender, exfoliative dermatosis. Despite treatment with high-dose steroids and intravenous immunoglobulin, the patient died. To the best of our knowledge, this is the first case of TEN associated with the use of sintilimab followed by lenvatinib. Early diagnosis and treatment of possibly fatal TEN reaction secondary to anti-PD-1 antibody therapy followed by lenvatinib is necessary.

Introduction

Toxic epidermal necrolysis (TEN) was first described by Alan Lyell in 1956. ¹ TEN is a rare adverse drug reaction characterized by skin involvement of >30% of the body surface area (BSA). ² Exfoliation relates to the extensive death of keratinocytes via apoptosis mediated by CD8+ T cells and natural killer cells, and it involves the cytotoxic secretory protein granulysin and interaction of the death receptor–ligand pair FAS-FasL. Cancers of the biliary system are highly aggressive tumors with a dismal prognosis, and only a few established options are available for systemic therapy. ³ Biliary tract cancer (BTC) describes a heterogeneous group of aggressive malignancies comprising intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma, gallbladder cancer, and ampulla of Vater cancer. Treatment concepts for BTC are genuinely influenced by precision and immune oncology. Although the development of immunotherapy has revolutionized the treatment landscape of several hematological and solid tumors, the role of immune checkpoint inhibitors (ICIs) and immune-based combinations in advanced BTC treatment remains unclear. ⁴ Novel combination strategies featuring ICIs should be considered with safety and pharmaceutical care in mind as these combination strategies become more popular. A crucial challenge is represented by the lack of validated predictive biomarkers for BTC. ⁵ Thus, it is difficult to identify patients who are likely to respond to immunotherapy. Adverse reactions to immunotherapy are unpredictable. Sintilimab is an anti-programmed cell death receptor-1 (PD-1) agent developed by Innovent Biologics in China and Eli Lilly Company in the US. Sintilimab is a fully human IgG4 monoclonal antibody that binds to PD-1, ⁶ thereby blocking the interaction of PD-1 with its ligands and helping to restore the endogenous antitumor T-cell response. Sintilimab has immune checkpoint-inhibitory and antineoplastic activities. Serious treatment-related adverse events such as pneumonitis and lung infection have been observed in 15% of patients. Other adverse events include infusion reaction, decreased platelet count, hyperthyroidism, liver function abnormality, peripheral neuropathy, and upper respiratory tract infection. Lenvatinib is the first-line treatment for intrahepatic cholangiocarcinoma. Its known adverse effects include hypertension, proteinuria, fatigue/asthenia, nausea, diarrhea, vomiting, stomatitis, and palmar–plantar erythrodysesthesia. Managing these adverse events is important during lenvatinib treatment. ⁷

To date, no cases of life-threatening cutaneous eruption, such as Stevens–Johnson syndrome (SJS)/TEN, have been reported in association with lenvatinib treatment. This report is the first to describe a case of TEN apparently related to treatment with sintilimab followed by lenvatinib.

Case report

Lenvatinib combined with sintilimab was administered to an otherwise healthy 78-year-old man with intrahepatic cholangiocarcinoma. The patient weighed 57 kg, and he had been physically fit in the past. The patient was married, and his spouse and children were in good health. He denied a family history of genetic disease or a history of similar diseases. We followed the CARE guidelines ⁸ and obtained consent from the patient from treatment.

Therefore, the patient was administered immunotherapy with sintilimab according to the standard schedule of 200 mg every 3 weeks, and first-line treatment with lenvatinib 8 mg every day was initiated the day after sintilimab therapy was started. After 18 days of oral lenvatinib administration and one dose of sintilimab, the patient developed a diffuse rash. While conducting the ward round, the chief physician discovered that the patient had developed erythema and blisters after immune and targeted therapy. The erythema and blisters gradually spread over his entire body. Some blisters ruptured and formed erosions. Based on the patient’s medical history obtained from an outside hospital, his medication history, and his symptoms, the diagnosis was toxic epidermal necrolysis and intrahepatic cholangiocarcinoma. The patient thought the symptoms were related to the recent administration of sintilimab or lenvatinib.

Diffuse purplish red patches were present over the patient’s body with a few scattered patches of normal skin. The patient exhibited multiple erythema and blisters on his face (Figure 1a) and trunk (Figure 1b–c) that gradually spread across his entire body (>30% BSA). Some of the blisters burst, and surface erosion was observed. The patient’s back (Figure 1d) and limbs were covered with scattered blisters and areas of surface erosion and exudation. Scabs were observed around the eyes, nose, and mouth. Thick, bloody scabs were observed on the lips and in the nasal passages. The external genitalia exhibited scattered areas of surface erosion (Figure 1e), and exudate was present. Fibroblast growth factor aerosol was applied externally to promote epidermal growth.

Figure 1.

(a) Multiple erythema and blisters on the face. (b–c) Multiple erythema and blisters on the trunk. (d) The patient’s back was covered with scattered blisters and areas of surface erosion and (e) External genitalia with scattered patches of surface erosion.

The patient complained of worsening oral mucosal pain and ocular symptoms. The TEN score was four. He had a body temperature of 39.5°C and general fatigue. Nikolsky’s sign was positive. Routine laboratory examinations revealed abnormal levels of aspartate aminotransferase (AST, 211.4 IU/L) and gamma-glutamyl transpeptidase (521.2 IU/L). Blood chemistry revealed increased C-reactive protein and glutamic oxalacetic transaminase levels. The patient had some protein in the urine and minor hypothyroidism (triiodothyronine, 0.21 ng/mL; thyroxine, 4.6 μg/dL; free T3, 1.45 pmol/L; free T4, 11.99 pmol/L; thyroid-stimulating hormone, 0.541 mIU/L). At the first examination, the following characteristics were noted: temperature, 39°C; pulse, 103 times/minute; respiratory rate, 20 times/minute; and blood pressure, 132/84 mmHg. Three days later, these characteristics had changed as follows: temperature, 36.9°C; pulse, 89 times/minute; respiratory rate, 21 times/minute; and blood pressure, 137/79 mmHg.

The patient’s total bilirubin level (TBIL) before death was 401.2 μmol/L (normal range, 3–22 μmol/L), and a progressive increase from 66.1 μmol/L to 168.6 μmol/L occurred 3 days after admission to our hospital. The abnormal TBIL level might have been related to the patient’s intrahepatic cholangiocarcinoma. Amylase was not measured. The patient’s creatinine level was 70.8 μmol/L on admission. After 1 week, the creatinine level increased to 128.5 μmol/L. Before the initiation of medication, the patient’s alanine aminotransferase (ALT), AST, and TBIL levels were 58 U/L, 87 U/L, and 51.2 μmol/L, respectively. The patient had poor liver function because of the underlying liver disease. After sintilimab and lenvatinib were administered for 1 day, the patient’s AST level increased to 133 U/L, whereas his ALT level increased slightly to 63 U/L. However, during this period, hepatoprotective drugs (polyene phosphatidylcholine capsule 456 mg tid) were administered simultaneously. His ALT levels gradually normalized, and his AST levels gradually decreased while remaining higher than the upper limit of normal. The gradual decrease in albumin content (from 36 g/L to 19.7 g/L) was related to the influence of the liver on protein synthesis, and it was considered that severe allergies could also aggravate the decrease in protein. The patient’s main manifestation of disease was limited to the skin, and organ damage was not observed. We found fecal occult blood positive. Based on these findings, the patient was diagnosed with TEN.

The patient was treated with 10 g/day gamma globulin for 2 days at another hospital and 25 g/day gamma globulin for 5 consecutive days at our hospital. The patient was treated with 80 mg/day methylprednisolone for 2 days at another hospital and 120 mg/day methylprednisolone for 7 days as steroid pulse therapy at our hospital. Thereafter, he was administered 60 mg/day methylprednisolone for 2 days. Proton pump inhibitors and hepatoprotective drugs were also used in treatment. Thirty-one days after anti-PD-1 treatment initiation and 14 days after the adverse reaction occurred, the patient died. The patient’s blood pressure dropped to 48/25 mmHg, and his creatinine increased to 128.5 μmol/L (normal range, 57–111 μmol/L).

A timeline with relevant data related to the treatment-related events is presented in Figure 2. The cause of the patient’s death was excessive effusion and hypovolemic shock.

Figure 2.

Timeline with relevant data regarding patient care.

In the present case, both the timeline and algorithm of drug causality for epidermal necrolysis (ALDEN) ⁹ score of 6 supported the probable role of sintilimab (Table 1). Lenvatinib (ALDEN score 2) could not be dismissed as the cause of the adverse reaction.

Table 1.

Algorithm of drug causality for epidermal necrolysis (ALDEN).

	Values
Criterion	Sintilimab	Lenvatinib
Delay from initial drug component intake to onset of reaction	+3	+3
Drug present in the body on index day	0	−1
Prechallenge/rechallenge	2	0
Dechallenge	0	0
Type of drug	2	0
Other cause	−1
Final score	6	2

<0, very unlikely; 0–1, unlikely; 2–3, possible; 4–5, probable; 6, very probable.

Discussion

BTC is usually diagnosed at a late stage, and it has a poor prognosis and only a few treatment options. Intrahepatic cholangiocarcinoma represents the second most common primary liver malignancy, accounting for 5% to 10% of all primary liver cancers. ¹⁰

Surgical resection remains the only potentially curative therapeutic option and the central cornerstone of treatment for intrahepatic cholangiocarcinoma. Adjuvant systemic treatment is recommended after resection or in the palliative setting. In addition, adjuvant and neoadjuvant chemotherapies as well as targeted therapies and immunotherapies based on molecular profiles can be applied. ¹¹

PD-1/PD-L1 inhibitors have displayed dramatic clinical efficacy in the treatment of multiple types of solid tumors, and the newly updated National Comprehensive Cancer Network guidelines recommend PD-1 blockade for advanced cholangiocarcinoma. The only FDA-approved immunotherapy for cholangiocarcinoma is pembrolizumab, an anti-PD-1 antibody. ¹² Zhang et al. treated a patient with metastatic intrahepatic cholangiocarcinoma with a PD-1 inhibitor. ¹³ The KEYNOTE-158 and KEYNOTE-028 trials assessed the efficacy and safety of pembrolizumab in the treatment of advanced biliary cancer. Grade 3 to 5 treatment-related adverse events occurred in 13.5% of patients in the KEYNOTE-158 trial (no grade 4; grade 5 renal failure, n = 1) and in 16.7% of patients in the KEYNOTE-028 trial (no grade 4/5). ¹⁴ Pembrolizumab provides durable antitumor activity in 6% to 13% of patients with advanced BTC regardless of PD-L1 expression, and it has manageable toxicity. Patients enrolled in the KEYNOTE-158 and KEYNOTE-028 trials experienced immune-mediated adverse events, including hypothyroidism, severe skin reaction, colitis, pneumonitis, hepatitis, hyperthyroidism, myositis, thyroiditis, and type 1 diabetes mellitus. Most events were mild-to-moderate in severity.

SJS is a type IV hypersensitivity reaction, mainly to drugs, which presents as mucocutaneous blistering and sloughing, and it can have a devastating clinical course. Many different drug classes include agents that have been implicated as causes of SJS. For example, allopurinol, carbamazepine, tyrosine kinase inhibitors, and ICIs can cause SJS. Although its incidence is approximately 1 to 2 cases per million/year, the mortality rate of SJS can be as high as 30% when the condition progresses to TEN. It is especially relevant considering that the use of such drugs in oncology is becoming more widespread for an ever-increasing number of cancer types. It has recently been reported that up to 22% of patients receiving anti-PD1 therapy develop inflammatory skin lesions ranging from mild maculopapular lesions to SJS-like rashes.

To date, no cases of SJS and TEN during treatment with either lenvatinib alone or in combination with anti-PD1 (sintilimab) therapy have been described. Additionally, no cases of severe cutaneous adverse reactions during anti-PD-1 therapy (sintilimab) alone have been reported.

After the first administration of sintilimab followed by lenvatinib for 18 days, the patient was admitted with fever, fatigue, and oral mucositis with severe pain. Blood chemistry analysis revealed increases in C-reactive protein and glutamic oxalacetic transaminase levels.

Sintilimab is a fully human IgG4 monoclonal antibody that binds to PD-1, thereby blocking the interaction of PD-1 with its ligands and helping to restore the endogenous antitumor T-cell response. Sintilimab was jointly developed by Innovent Biologics and Eli Lilly Company and was approved in China for the treatment of classical Hodgkin’s lymphoma for patients who experience relapse or refractory disease after ≥2 lines of systemic chemotherapy and for patients with non-small cell lung cancer. Sintilimab is being evaluated in phase I, II, and III trials for the treatment of various solid tumors, including hepatocellular carcinoma. ¹⁵ Phase I/II development of sintilimab for the treatment of solid tumors is underway in the US, and the US FDA accepted an Investigational New Drug application for sintilimab in January 2018.

Sintilimab exhibited linear pharmacokinetics over a 1 to 10 mg/kg dose range. Its half-life is 19.6 days. Sintilimab blocks PD-1 from interacting with its ligands (PD-L1 and PL-L2). Serious treatment-related adverse events, such as pneumonitis and lung infection, occurred in 15% of patients treated with sintilimab. Other adverse events included infusion reaction, decreased platelet counts, hyperthyroidism, liver function abnormality, peripheral neuropathy, and upper respiratory tract infection. ¹⁶ No treatment-emergent adverse events resulting in death occurred. Different scholars have described cases of drug-induced SJS/TEN (Table 2).

Table 2.

Severe cutaneous adverse reactions attributable to different anticancer drugs.

Ref.	Type of report	Drug inducing adverse reactions	Disease	Severity of reaction	Outcome
16	Case report	Nivolumab	NSCLC	SJS	Improved
17	Case report	Pembrolizumab	Lung cancer	SJS	Improved
18	Case report	Nivolumab in a patient treated with ipilimumab	Metastatic melanoma	TEN	Improved, died after 4 months
19	Case report	Nivolumab	Metastatic melanoma	TEN	Died
22	Case report	Crizotinib	NSCLC	TEN	Died
23	Case report	Pemetrexed plus cisplatin and gefitinib	NSCLC	TEN	Improved

NSCLC, non-small cell lung cancer; SJS Stevens–Johnson syndrome; TEN, toxic epidermal necrolysis.

Salati et al. reported a case of SJS arising during nivolumab treatment for NSCLC. ¹⁷ Haratake et al. presented a case of SJS induced by pembrolizumab in a patient with lung cancer. ¹⁸

Further supporting the development of TEN following anti-PD-1 therapy, Nayar et al. reported a case of a toxic epidermal necrolysis-like reaction associated with nivolumab in a patient with ipilimumab-refractory metastatic melanoma. ¹⁹ The patient received extended rehabilitation but died 4 months after the diagnosis of TEN because of disease progression and sepsis.

Vivar et al. presented a case of a patient with a fatal toxic epidermal necrolysis eruption associated with the use of nivolumab for the treatment of metastatic melanoma. ²⁰ Despite treatment with infliximab, high-dose steroids, and immunoglobulin, the patient died. In this study, we described to our knowledge the first case of TEN associated with sintilimab reported in the anti-PD1 therapy literature. There is growing evidence of the ability of anti-PD-1 antibodies to induce TEN. Cutaneous eruption should be considered with a high degree of clinical suspicion during immunotherapy.

Lenvatinib is a small-molecule tyrosine kinase inhibitor that inhibits vascular endothelial growth factor receptor, fibroblast growth factor receptor, platelet-derived growth factor receptor α, stem cell factor receptor, and rearranged during transfection. Lenvatinib is a promising drug that has displayed therapeutic effects against various solid tumors.

Lenvatinib is the first-line treatment for intrahepatic cholangiocarcinoma. Its known adverse effects include hypertension, proteinuria, fatigue/asthenia, nausea, diarrhea, vomiting, stomatitis, and palmar-plantar erythrodysesthesia. ²¹ Managing these adverse events is also important during lenvatinib treatment. ²² In large clinical trials of lenvatinib, elevations in serum aminotransferase levels were common, occurring in 52% of patients.

Yang et al. presented the case of a 75-year-old Chinese man with advanced NSCLC with ALK fusion ²³ who developed TEN after 56 days of crizotinib treatment and died from adverse events. The occurrence of severe cutaneous necrolysis involving the skin and mucous membranes during tyrosine kinase inhibitor treatment should alert clinicians to be aware of TEN and take prompt actions.

Huang et al. reported a rare presentation of TEN involving adverse effects of pemetrexed plus cisplatin and gefitinib combination treatment in a 42-year-old woman diagnosed with metastatic NSCLC harboring an endothelial growth factor receptor (EGFR) mutation. ²⁴

Overall, small-molecule tyrosine kinase inhibitors are associated with TEN. TEN is a rare acute life-threatening mucocutaneous disorder that is mostly drug-related (80%–95%).

To date, no case of life-threatening cutaneous eruption, such as SJS/TEN, has been reported in patients treated with sintilimab and lenvatinib. This report is the first description of a case of SJS/TEN apparently related to sintilimab followed by lenvatinib. We obtained new knowledge of the toxicity and future prospects of lenvatinib therapy in hepatocellular carcinoma.

The patient experienced a positive reaction after approximately 18 days of oral lenvatinib therapy after sintilimab treatment. This result suggested that both lenvatinib and sintilimab could have caused the adverse reaction.

An increasing number of targeted agents have been reported to induce life-threatening severe cutaneous adverse reactions, including SJS/TEN, drug-associated rash with eosinophilia, and systemic symptoms/drug-induced hypersensitivity syndrome. ²⁵ For example, erlotinib is a tyrosine kinase inhibitor (TKI) specific to EGFR. Its latency (median range) of severe cutaneous adverse reactions is 8 (8).

Gefitinib is an EGFR-TKI. Its latency (median range) of severe cutaneous adverse reactions is 8 (7–10). Nivolumab and pembrolizumab are PD-1 inhibitors. The latency (median range) of severe cutaneous adverse reactions of nivolumab is 64.5 (39–90), and that for pembrolizumab is 83.5 (7–140). We found that the latency of some PD-1 inhibitors was longer than that of EGFR-TKIs.

According to the patient’s medication history and the time of onset of skin lesions, sintilimab is considered the most likely cause of the adverse reaction. This drug can cause severe skin allergies. However, the incidence is low, and the latency of severe allergies to PD-1 is longer than that reported for TKIs, in line with the timeline of the patient’s adverse reaction. The drug has a half-life of 19.6 days, and the drug is slowly eliminated. Lenvatinib-related TEN and SJS have not been reported in the literature, but other TKIs have been reported to cause severe skin allergies. Therefore, lenvatinib cannot be dismissed as the cause of the adverse reaction. The patient’s reaction was severe, and his prognosis was poor. In the present case, the ALDEN score supported the probable role of sintilimab and the possible role of lenvatinib.

Anticancer drugs, including chemotherapy, targeted therapy, and recent immunotherapy, cause skin reactions ranging from mild skin rash to life-threatening severe cutaneous adverse reactions. Some researchers have demonstrated an upregulation of the expression of major inflammatory cytokines through gene expression analysis of TEN-like lesional skin from anti-PD-1–treated patients. ²⁶

Tangamornsuksan et al. found that the HLA phenotype is significant in drug allergy and toxicity, and therefore, we plan to test the HLA phenotype in patients with this drug eruption, study the relationship between HLA and allergy, and use HLA to determine whether the patient is allergic to this drug, which has guiding significance in avoiding drug hypersensitivity.^27–28 Steroids should be administered as soon as possible after the diagnosis of immune-related adverse events. ²⁹ Intravenous immunoglobulin should be administered as soon as possible after the diagnosis of TEN. ³⁰

The limitations of this case study included a lack of examination of tissues and HLA gene analysis.

We aimed to provide a case report of SJS/TEN associated with sintilimab and lenvatinib to increase clinician knowledge and drive future studies on patient management.

Treatment with ICIs, whether as monotherapy or as combination therapy that includes other anticancer agents, has yielded controversial results to date. Most patients experience disappointing clinical outcomes. The available data on the predictors of patient response to ICIs for BTC are conflicting, and no single biomarker is available for the selection of patients likely to benefit from this therapeutic approach.

The primary lesson of this case report is that full pharmaceutical care must be provided during the course of immunotherapy and targeted therapy, and when a patient experiences a possible adverse reaction, they should seek timely medical treatment.

Data availability statement

All data generated or analyzed during this study are included in this published article and are available from the corresponding author upon reasonable request.

Ethics statement

Because the patient died, we did not obtain written informed consent to publish this report, and the requirement for consent was waived. We have de-identified the details such that the identity of the patient may not be ascertained in any way. The requirement for ethics board approval was waived because this was a case report.