Abstract
Lenvatinib has anti-tumor activity against advanced hepatocellular carcinoma (HCC). Hypothyroidism is also a frequent complication in patients treated with lenvatinib. However, studies on lenvatinib-induced thyroid toxicity and destructive thyroiditis are limited. Therefore, this study aimed to clarify the frequency and timing of thyroid abnormalities in lenvatinib for unresectable HCC. This retrospective study enrolled 50 patients with advanced HCC treated with lenvatinib. Patients were classified to have euthyroid, subclinical hypothyroidism, overt hypothyroidism, and thyrotoxicosis. The timing of thyroid dysfunction was assessed, and risk factors for incident hypothyroidism or thyrotoxicosis were evaluated using multivariate models. Subclinical hypothyroidism, overt hypothyroidism, and thyrotoxicosis occurred in 7 (14.0%), 26 (52.0%), and 5 (10.0%) patients, respectively. In the 33 patients with hypothyroidism, 27 (84.4%) developed the condition within 2 weeks of starting lenvatinib treatment. Of the 5 patients with thyrotoxicosis, 3 developed the condition within 8 weeks of starting lenvatinib administration. One patient developed thyrotoxicosis in only 1 week of the initiation of treatment. No correlation between the presence of antibodies and the incidence and severity of thyroid dysfunction due to the autoimmune mechanism was observed. The progression-free survival was significantly better in the hypothyroidism group. Lenvatinib treatment for unresectable HCC not only causes hypothyroidism, but also thyrotoxicosis. Moreover, these thyroid conditions develop within the early period of treatment at a higher prevalence. Patients with thyroid dysfunction had better prognosis. Based on these results, in patients administered with lenvatinib, there is need for careful assessment for the possibility of thyroid dysfunction from the onset of treatment.
LENVATINIB is an oral angiogenesis inhibitor that targets vascular endothelial growth factor (VEGF) receptors 1–3, fibroblast growth factor receptors (FGFR) 1–4, platelet-derived growth factor receptor alpha, and the RET and KIT oncogenes [1-4]. Particularly, the strong inhibitory activity of FGFR4 is considered to be effective for suppressing invasive metastasis of hepatocellular carcinoma (HCC); and lenvatinib has been reported to be non-inferior to sorafenib with regard to overall survival in cases of advanced HCC [5, 6].
However, several recent studies have reported tyrosine kinase inhibitor (TKI)-induced hypothyroidism and thyroid dysfunction [7-9]. The most common reported adverse events (AEs) with lenvatinib treatment for HCC are hypertension, hand-foot syndrome, decreased appetite, proteinuria, and fatigue [5, 6] Hypothyroidism has been reported in 16% [6] and 21.7% [5] of patients treated with lenvatinib, but the time of onset of hypothyroidism and the incidence of lenvatinib-induced thyrotoxicosis are unclear. In addition, the risk factors of hypothyroidism and thyrotoxicosis due to lenvatinib treatment for unresectable HCC are unknown.
In this study, we aimed to clarify the frequency and timing of thyroid abnormalities in lenvatinib treatment for unresectable HCC. Toward this goal, we retrospectively collected data and compared the incidence and time of onset of thyroid dysfunction in Japanese patients with unresectable HCC treated with lenvatinib.
Methods
Patients
Written informed consent was obtained from all study participants before enrollment, and all study protocols were approved by the institutional ethics committee (permit number: 1901004). This study was conducted after registration with the University Hospital Medical Information Network (UMIN trial number: UMIN000036397). Fifty patients with unresectable HCC treated with lenvatinib at Ehime University Hospital, Ehime Prefectural Central Hospital, and Matsuyama Red Cross Hospital in Japan between April 2018 to November 2018 were enrolled in this study. The inclusion criteria were (1) one or more measurable target lesions based on the modified Response Evaluation Criteria in Solid Tumours (mRECIST) [10], (2) Barcelona Clinic Liver Cancer stage B or C categorization [11], (3) age ≥20 years, (4) Child-Pugh class A or B, and an (5) Eastern Cooperative Oncology Group performance status score of 0 or 1. The exclusion criterion was history of other thyroid diseases. Patients were classified as euthyroid (normal thyroid-stimulating hormone [TSH]); subclinical hypothyroidism (TSH, 5–10 mIU/L if the free thyroxine (FT4) was normal); overt hypothyroidism (TSH, >10 mIU/L, low FT4 or requiring thyroid hormone replacement); and thyrotoxicosis (TSH <0.5 mIU/mL and either free triiodothyronine (FT3) >4.0 pg/mL or FT4 >1.7 ng/dL).
Data collection
The data collected on thyroid function included history of thyroid disease; treatment with thyroid hormone or antithyroid drugs; serum TSH, FT3, and FT4 levels; anti-human TSH receptor antibody; anti-thyroglobulin antibody (TgAb); antithyroid peroxidase antibody (TPOAb); and date of first occurrence of lenvatinib-related thyroid abnormality.
Lenvatinib treatment and adverse events
Patients received oral lenvatinib 12 mg/day or 8 mg/day for a bodyweight of ≥60 kg or <60 kg, respectively. Dose interruptions, followed by dose reductions for lenvatinib-related toxicities (to 8 mg and 4 mg/day, or 4 mg every other day) were permitted. According to the guideline for lenvatinib administration, the drug dose should either be reduced or the treatment interrupted when a patient develops ≥grade 3 severe AEs or any unacceptable grade 2 drug-related AEs. AEs were assessed according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 4.0 [12]. Patients diagnosed with lenvatinib-induced overt hypothyroidism were given thyroid hormone replacement therapy.
Follow-up
Patients were followed-up every two weeks from the start of lenvatinib treatment. When progressive disease (PD) was confirmed, the treatment was discontinued. In addition, follow-up was also stopped at the time of the treatment was discontinued, either for side effects or deterioration of the general condition. Patients were instructed to visit any of the three participating hospitals for any physical change. Blood test (liver function test, thyroid function test, alpha-fetoprotein, and des-γ-carboxy prothrombin) was performed every 2 weeks. Contrast-enhanced computerized tomography was performed every month, and treatment efficacy was evaluated according to mRECIST [10].
Statistical analysis
Data were analyzed using the Student t test for unpaired data and the chi-square test and Fisher’s exact test as appropriate. The Pearson product-moment correlation coefficient was used to determine the correlation between variables, and a backward stepwise procedure was used to select the independent prognostic variables. The Kaplan-Meier method was used for the survival curve. Multivariate logistic regression was performed using the Wald test. All data were analyzed using JMP software (version 13; SAS Institute Japan, Tokyo, Japan).
Results
Patients
In total, 50 patients met the inclusion criteria. Their baseline characteristics are shown in Table 1. Subclinical hypothyroidism, overt hypothyroidism, and thyrotoxicosis occurred in 7 (14.0%), 26 (52.0%), and 5 (10.0%) patients, respectively.
Table 1
Patient baseline characteristics
| Characteristics |
(n = 50) |
| Age (years) |
70.9 ± 9.5 |
| Sex (M/F) |
50 (39/11) |
| AST (U/L) |
50.8 ± 27.5 |
| ALT (U/L) |
39.3 ± 32.5 |
| Serum albumin (g/dL) |
3.6 ± 0.5 |
| Platelet count (×104/μL) |
13.3 ± 6.1 |
| Prothrombin time (%) |
86.2 ± 13.4 |
| Total bilirubin (mg/dL) |
0.9 ± 0.4 |
| Baseline TSH (μIU/mL) |
3.6 ± 1.9 |
| Baseline FT3 (pg/mL) |
2.4 ± 0.7 |
| Baseline FT4 (ng/mL) |
1.4 ± 0.5 |
| Baseline TPOAb (IU/mL) |
14.4 ± 8.9 |
| Baseline TgAb (IU/mL) |
48.7 ± 96.9 |
| AFP (ng/mL) |
10,951.2 ± 52,557.0 |
| DCP (mAU/mL) |
10,739.5 ± 37,133.6 |
| Etiology (HCV/HBV/NAFLD (NASH)/alcohol/Others) |
23/10/5/5/7 |
| Child-Pugh class (A/B/C) |
45/5/0 |
| Intrahepatic tumor size (cm) |
3.2 ± 2.4 |
| Number of intrahepatic tumors (none/single/multiple) |
3/5/42 |
| TNM stage (I/II/III/IV) |
1/6/14/29 |
| Thyroid abnormalities (euthyroid/subclinical hypothyroidism/overt hypothyroidism/thyrotoxicosis) |
12/7/26/5 |
| Initial lenvatinib dose (8 mg/12 mg) |
23/27 |
| Median observation period after treatment (days) |
127 (IQR: 85.8–145.5) |
Data are expressed as mean ± standard deviation or n (%).
AST, aspartate aminotransferase; ALT, alanine aminotransferase; TSH, thyroid-stimulating hormone; FT3, free triiodothyronine; FT4, free thyroxine; TPOAb, antithyroid peroxidase antibody; TgAb, anti-thyroglobulin antibody; HCV, hepatitis C virus; HBV, hepatitis B virus; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; AFP, alpha-fetoprotein; DCP, des-γ-carboxy prothrombin; TNM, tumor–node–metastasis; IQR, interquartile range.
Subclinical hypothyroidism occurred at 2, 4, and 8 weeks after the start of treatment in 3 (42.9%), 3 (42.9%), and 1 (14.3%) patient, respectively. Overt hypothyroidism occurred at 2, 4, and 8 weeks after the start of treatment in 22 (84.6%), 4 (15.4%), and 0 (0%) patient, respectively. Multivariate analysis showed no correlation in female sex, thyroid hormone, and thyroid autoantibody with the occurrence of hypothyroidism (data not shown).
Lenvatinib-induced thyrotoxicosis
In the 5 patients who developed thyrotoxicosis, the condition occurred at 2, 4, and 8 weeks after the start of treatment in 1, 1, and 3 patients, respectively. In all 5 patients, the TSH, free T3, and free T4 spontaneously reverted to normal within 6 to 13 weeks after dose reduction or treatment discontinuation. Furthermore, the TgAb and TPOAb titers did not change throughout the treatment courses. In 1 patient who developed thyrotoxicosis at 1 week after the start of administration, laboratory test results also revealed positivity for anti-thyroid antibodies (TgAb, 129.0 IU/mL [normal range, 0–27.9]) and negativity for other antibodies (TPOAb, 15.5 IU/mL [normal range 0–15.9]; anti-human TSH receptor antibody, <0.3 IU/L [normal range 0–2.0]; and thyroid-stimulating antibody, 109% [normal range <120]). In addition, 99m-technetium (Tc) scintigraphy demonstrated a markedly decreased (0.12%) radioactive uptake. Thus, the thyroid dysfunction in this patient was diagnosed as lenvatinib-induced destructive thyroiditis. Notably, multivariate analysis showed that thyroid autoantibodies (e.g., TPOAb and TgAb) were not risk factors for thyrotoxicosis (Table 2).
Table 2
Multivariate analysis of factors contributing to thyrotoxicosis
|
Odds ratio |
95% CI |
p value |
| Baseline TSH |
0.799 |
0.174–3.680 |
0.7738 |
| Baseline FT3 |
1.047 |
0.084–4.434 |
0.5423 |
| Baseline FT4 |
1.545 |
0.045–5.310 |
0.2335 |
| Baseline TgAb |
1.153 |
0.769–1.728 |
0.4915 |
| Baseline TPOAb |
0.992 |
0.960–1.025 |
0.6353 |
| Baseline TSAb |
0.289 |
0.305–1.101 |
0.5968 |
CI, confidence interval; TSH, thyroid-stimulating hormone; FT3, free triiodothyronine; FT4, free thyroxine; TPOAb, antithyroid peroxidase antibody; TgAb, anti-thyroglobulin antibody; TSAb, thyroid-stimulating antibody.
Treatment efficacy was evaluated according to mRECIST. There were 3 (6.0%), 13 (26%), 25 (50.0%), and 9 (18.0%) cases of complete response, partial response, stable disease, and PD, respectively. Treatment was completed in 26 patients (52.0%), while it was stopped in 5 (10.0%) and 15 (30.0%) due to PD and intolerable side effects. However, treatment was discontinued in 4 (8.0%) patients due to deteriorating general condition. The median treatment period was 55 days (range, 15–180 days), and the median observation period after the start of lenvatinib treatment was 127 days. Comparison of progression-free survival between the hypothyroidism group and the non-hypothyroidism group showed that the hypothyroidism group had a significantly longer period of tumor progression (Fig. 1). However, there was no significant difference in progression-free survival between the thyrotoxicosis group and non-thyrotoxicosis group.
Adverse events of lenvatinib
The AE of lenvatinib are shown in Table 3. No G4 AE occurred during the observation period. Hypothyroidism was the most common AE (all grades, n = 33; G3, n = 1), followed by general fatigue, appetite loss, and hoarseness. Four patients died during the observation period; 3 died of liver cancer, and 1 died of hepatic failure. One patient with stage IVb HCC at the start of lenvatinib stopped treatment because of severe AE (hepatic coma) on day 9, and the patient died due to hepatic failure on day 20.
Table 3
Lenvatinib-related adverse events
|
Grade 1 or 2 |
Grade 3 |
All grades |
| Hypothyroidism |
32 |
1 |
33 |
| General fatigue |
22 |
1 |
24 |
| Appetite loss |
14 |
3 |
17 |
| Hoarseness |
11 |
0 |
11 |
| HFS |
4 |
3 |
7 |
| Diarrhea |
5 |
2 |
7 |
| Hypertension |
5 |
2 |
7 |
| Urine protein |
5 |
1 |
6 |
| Body weight loss |
5 |
0 |
5 |
| Thyrotoxicosis |
3 |
2 |
5 |
| Decreased platelets |
2 |
1 |
3 |
| Dysgeusia |
2 |
0 |
2 |
| Fever |
1 |
0 |
1 |
| Rash |
1 |
0 |
1 |
| Hepatic coma |
— |
1 |
1 |
HFS, hand-foot syndrome; —, no setting for the applicable grade in the National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 4.0.
Discussion
Lenvatinib is approved for unresectable HCC. Thyroid dysfunction in TKI treatment is a commonly recognized AE. Axitinib, pazopanib, sorafenib, sunitinib, and lenvatinib have been reported to cause thyroid-related AEs [5, 6, 13-15]. Lenvatinib has been used for thyroid cancer and thyroidectomy; however, reports of AEs of thyroid dysfunction are limited.
The incidence of lenvatinib-induced hypothyroidism has been reported to be 16%–21.7% [5, 6]. Although the number of patients we reported here was small, the incidence of thyroid function abnormalities was quite high relative to these earlier reports. In this study of 50 patients with unresectable HCC who have euthyroid at baseline, 66% developed hypothyroidism (overt hypothyroidism: 52%) and 10% developed thyrotoxicosis during lenvatinib treatment. The total number of hypothyroidism cases may have increased because subclinical hypothyroidism was included. From these results, in lenvatinib-administered cases, we need to be careful about the onset of thyroid dysfunction from early after administration start, especially in Japanese patients with unresectable hepatocellular carcinoma.
The mechanism by which TKIs induce thyroiditis is unclear. Several studies reported that TKIs induce hypothyroidism through various mechanisms, such as destructive thyroiditis and tissue ischemia [16, 17]. Particularly, one potential mechanism is that TKIs cause apoptosis of the thyroid follicular cells, leading to destructive thyroiditis. Second, the prevention of VEGF binding to normal thyroid cells or inhibition of thyroid blood flow can cause destruction. Third, an as-yet-undescribed autoimmune mechanism affecting the thyroid function may result in thyrotoxicosis. Thyrotoxicosis may have occurred in the present patients by any of these mechanisms [18].
The time of onset of hypothyroidism is also unclear. One study reported that the TSH gradually increases at an early stage from the start of lenvatinib treatment [19]. Also in the current study, 84.6% of overt hypothyroidism cases developed in 2 weeks from the start of lenvatinib treatment. Thus, thyroid function should be frequently assessed starting from the early period of lenvatinib treatment.
There were five cases of lenvatinib-induced thyrotoxicosis, but TSH, free T3, and free T4 spontaneously reverted to normal after dose reduction or treatment discontinuation. In the hyperthyroid state, only beta-blockers should be administered. When hypothyroidism is symptomatic (ex: fatigue and appetite loss), L-thyroxine should be administered as soon as possible. No independent contributing factors to develop thyrotoxicosis were detected in this study. Thus, presumptive prediction of thyrotoxicosis is considered difficult in the present study.
Interestingly, progression-free survival was better in cases with lenvatinib-induced hypothyroidism. This may suggest that VEGF inhibition occurred more strongly in cases with hypothyroidism [5]. Furthermore, there are several reports that the thyroid hormone is involved in the activation of FGFR [20-23], and lenvatinib has a strong inhibitory activity on FGFR [6]. Therefore, lenvatinib may be involved both in the onset of hypothyroidism and the long-term progression-free survival. If VEGF and FGFR are inhibited in the thyroid, molecules in hepatic malignancies may also be inhibited. There is limited reporting on the association between hypothyroidism and progression-free survival in the treatment of lenvatinib for thyroid cancer. Lenvatinib is administered to thyroid cancer patients, and there are generally no thyroid tissues in such patients. Therefore, the involvement of thyroid hormone in FGFR activation may be related to the development of good progression-free survival in patients with hypothyroidism during lenvatinib treatment for unresectable HCC.
In this study, there were 9 (18.0%) cases of PD with lenvatinib treatment. In patients with PD who were treated with lenvatinib as first line treatment, treatment with sorafenib is considered as the second line. Patients, who have already received sorafenib treatment as the first line, are considered to have been treated with transarterial chemoembolization or hepatic arterial infusion chemotherapy [24].
This study had several limitations. First, the number of cases was limited. Investigations with a higher number of cases are needed to verify the mechanisms of thyroid dysfunction in lenvatinib treatment. Secondly, because this was a retrospective study, the methodology for ultrasonographic examination of the thyroid gland was not uniform and not all patients underwent the examination.
In conclusion, lenvatinib is an effective treatment for unresectable HCC and may improve its prognosis. The occurrence of hypothyroidism and thyrotoxicosis may have been associated with treatment-induced injury of the thyroid gland. Despite the high number of treatment-related AEs, all cases were managed via dose modification and medical therapy.
Abbreviations
AE, adverse event; FGFR, fibroblast growth factor receptors; HCC, hepatocellular carcinoma; mRECIST, modified Response Evaluation Criteria in Solid Tumors; PD, progressive disease; PDGF, platelet-derived growth factor; TgAb, anti-thyroglobulin antibody; TKI, tyrosine kinase inhibitor; TPOAb, antithyroid peroxidase antibody; UMIN, University Hospital Medical Information Network; VEGF, vascular endothelial growth factor
Disclosure
None of the authors have any potential conflicts of interest associated with this research.
Author disclosure statement: No competing financial interests exist.
References
- 1 Matsui J, Yamamoto Y, Funahashi Y, Tsuruoka A, Watanabe T, et al. (2008) E7080, a novel inhibitor that targets multiple kinases, has potent antitumor activities against stem cell factor producing human small cell lung cancer H146, based on angiogenesis inhibition. Int J Cancer 122: 664–671.
- 2 Matsui J, Funahashi Y, Uenaka T, Watanabe T, Tsuruoka A, et al. (2008) Multi-kinase inhibitor E7080 suppresses lymph node and lung metastases of human mammary breast tumor MDA-MB-231 via inhibition of vascular endothelial growth factor-receptor (VEGF-R) 2 and VEGF-R3 kinase. Clin Cancer Res 14: 5459–5465.
- 3 Tohyama O, Matsui J, Kodama K, Hata-Sugi N, Kimura T, et al. (2014) Antitumor activity of lenvatinib (e7080): an angiogenesis inhibitor that targets multiple receptor tyrosine kinases in preclinical human thyroid cancer models. J Thyroid Res 2014: 638747.
- 4 Yamamoto Y, Matsui J, Matsushima T, Obaishi H, Miyazaki K, et al. (2014) Lenvatinib, an angiogenesis inhibitor targeting VEGFR/FGFR, shows broad antitumor activity in human tumor xenograft models associated with microvessel density and pericyte coverage. Vasc Cell 6: 18.
- 5 Ikeda K, Kudo M, Kawazoe S, Osaki Y, Ikeda M, et al. (2017) Phase 2 study of lenvatinib in patients with advanced hepatocellular carcinoma. J Gastroenterol 52: 512–519.
- 6 Kudo M, Finn RS, Qin S, Han KH, Ikeda K, et al. (2018) Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet 391: 1163–1173.
- 7 Rini BI, Escudier B, Tomczak P, Kaprin A, Szczylik C, et al. (2011) Comparative effectiveness of axitinib versus sorafenib in advanced renal cell carcinoma (AXIS): a randomised phase 3 trial. Lancet 378: 1931–1939.
- 8 Smith MR, Sweeney CJ, Corn PG, Rathkopf DE, Smith DC, et al. (2014) Cabozantinib in chemotherapy-pretreated metastatic castration-resistant prostate cancer: results of a phase II nonrandomized expansion study. J Clin Oncol 32: 3391–3399.
- 9 Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, et al. (2009) Overall survival and updated results for sunitinib compared with interferon alfa in patients with metastatic renal cell carcinoma. J Clin Oncol 27: 3584–3590.
- 10 Lencioni R, Llovet JM (2010) Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis 30: 52–60.
- 11 Bruix J, Sherman M; American Association for the Study of Liver Diseases (2011) Management of hepatocellular carcinoma: an update. Hepatology 53: 1020–1022.
- 12 National Cancer Institute, Division of Cancer Treated and Diagnosis Cancer Therapy Evaluation Program. Adverseevents/CTCAE. Available at https://ctep.cancer.gov/protocolDevelopment/electronic_applications/ctc.htm#ctc_40. Accessed May 31, 2018.
- 13 Ohba K, Takayama T, Matsunaga H, Matsushita A, Sasaki S, et al. (2013) Inappropriate elevation of serum thyrotropin levels in patients treated with axitinib. Thyroid 23: 443–448.
- 14 Motzer RJ, Hutson TE, Cella D, Reeves J, Hawkins R, et al. (2013) Pazopanib versus sunitinib in metastatic renal-cell carcinoma. N Engl J Med 369: 722–731.
- 15 van Doorn L, Eskens FA, Visser TJ, van der Lugt A, Mathijssen RH, et al. (2011) Sorafenib induced thyroiditis in two patients with hepatocellular carcinoma. Thyroid 21: 197–202.
- 16 Illouz F, Braun D, Briet C, Schweizer U, Rodien P (2014) Endocrine side-effects of anti-cancer drugs: thyroid effects of tyrosine kinase inhibitors. Eur J Endocrinol 171: R91–R99.
- 17 Ahmadieh H, Salti I (2013) Tyrosine kinase inhibitors induced thyroid dysfunction: a review of its incidence, pathophysiology, clinical relevance, and treatment. Biomed Res Int 2013: 725410.
- 18 Hirooka M, Ochi H, Hiraoka A, Koizumi Y, Matsuura B, et al. (2019) Destructive thyroiditis induced by lenvatinib in three patients with hepatocellular carcinoma. Intern Med 58: 791–795.
- 19 Sato S, Muraishi K, Tani J, Sasaki Y, Tokubuchi I, et al. (2010) Clinical characteristics of thyroid abnormalities induced by sunitinib treatment in Japanese patients with renal cell carcinoma. Endocr J 57: 873–880.
- 20 Stevens DA, Harvey CB, Scott AJ, O’Shea PJ, Barnard JC, et al. (2003) Thyroid hormone activates fibroblast growth factor receptor-1 in bone. Mol Endocrinol 17: 1751–1766.
- 21 Barnard JC, Williams AJ, Rabier B, Chassande O, Samarut J, et al. (2005) Thyroid hormones regulate fibroblast growth factor receptor signaling during chondrogenesis. Endocrinology 146: 5568–5580.
- 22 St Bernard R, Zheng L, Liu W, Winer D, Asa SL, et al. (2005) Fibroblast growth factor receptors as molecular targets in thyroid carcinoma. Endocrinology 146: 1145–1153.
- 23 Szarama KB, Gavara N, Petralia RS, Chadwick RS, Kelley MW (2013) Thyroid hormone increases fibroblast growth factor receptor expression and disrupts cell mechanics in the developing organ of corti. BMC Dev Biol 13: 6.
- 24 Llovet JM, Montal R, Sia D, Finn RS (2018) Molecular therapies and precision medicine for hepatocellular carcinoma. Nat Rev Clin Oncol 15: 599–616.
