Treatment for Advanced Hepatocellular Carcinoma: Current Standard and the Future

Watch a video presentation of this article

Abbreviations

AE

adverse event

BCLC

Barcelona Clinic Liver Cancer

BRISK‐PS

brivanib study in HCC patients at risk post sorafenib

CI

confidence interval

CRT

conformal radiotherapy

EHS

extrahepatic spread

EVOLVE‐1

the first everolimus for liver cancer evaluation

FDA

US Food and Drug Administration

HAIC

hepatic arterial infusion chemotherapy

HCC

hepatocellular carcinoma

HR

hazard ratio

LFP

5‐fluorouracil and cisplatin

MVI

macrovascular invasion

NA

not available

NS

not significant

ORR

overall response rate

OS

overall survival

PD‐1

programmed death receptor‐1

PFS

progression‐free survival

RCT

randomized controlled trial

RT

external beam radiotherapy

SBRT

stereotactic body radiation therapy

SHARP

sorafenib hepaticellular carcinoma assessment randomized protocol

SIRT

selective internal radiation therapy

TACE

transarterial chemoembolization

TTF

time to treatment failure

TTP

time to progression

Hepatocellular carcinoma (HCC) is the third leading cause of death worldwide with increasing incidence and mortality in the United States.1, 2 High HCC‐associated mortality is in part due to the high proportion of patients diagnosed with advanced stage HCC and historical lack of effective systemic therapies for HCC.

HCC staging is unique because liver function and functional status, in addition to tumor burden, are integral determinants of stage and prognosis. Although staging systems vary, parameters that define advanced stage HCC eligible for therapy include presence of portal vein tumor invasion and/or extrahepatic metastases, with relatively preserved liver function and functional status. Generally, systemic therapy trials excluded patients with Child Pugh class B and C cirrhosis, largely because of the competing risk for mortality with cirrhosis. Thus, for many therapies, there are little data on efficacy and tolerability in patients with more advanced liver disease. Systemic therapies may also be appropriate in those patients with unresectable HCC who are not eligible for or are unlikely to benefit from locoregional therapies, although the decision on timing of when to initiate systemic therapy in a patient with intermediate HCC who is eligible for recurrent locoregional therapy remains an open question. In this review, we discuss contemporary approaches and ongoing studies for the treatment of patients with advanced HCC.

Systemic Therapy

Multikinase Inhibitors

Until recently, sorafenib has been the only US Food and Drug Administration (FDA)–approved first‐line agent for advanced HCC. Sorafenib has been associated with modest improvement in overall survival (OS) as compared with placebo in patients with Child‐Pugh A cirrhosis and an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 1 in both a trial setting (10.7 versus 7.9 months [hazard ratio (HR), 0.73; 95% confidence interval (CI), 0.58‐0.92] in the phase 3 sorafenib hepatocellular carcinoma assessment randomized protocol (SHARP) trial and 6.5 versus 4.2 months in the Asia‐Pacific trials, respectively)3 and in multiple real‐world observational cohorts that included patients with varying liver function.4, 5 Adverse events (AEs), including diarrhea, fatigue, and palmar‐plantar erythrodysesthesia, are frequent (80%) and led to drug discontinuation in approximately 20% of patients in the global investigation of therapeutic decisions in hepatocellular carcinoma and of its treatment with sorafenib (GIDEON) observational cohort study.4 Lenvatinib was recently shown to be noninferior to sorafenib (13.6 versus 12.3 months; HR, 0.92; 95% CI, 0.79‐1.06) in the REFLECT study with similar side effects (hypertension, diarrhea, fatigue, weight loss, palmar‐plantar erythrodysesthesia) and frequency of grade ≥3 AEs (75%), resulting in the recent first‐line approval of lenvatinib for HCC by the FDA.6 Compared with sorafenib, lenvatinib is also associated with higher rates of proteinuria (25%) and dysphonia (24%). Secondary endpoints of time to progression (HR, 0.60; 95% CI, 0.51‐0.71) and objective response (HR, 3.13; 95% CI, 2.15‐4.56) were superior in the lenvatinib arm; however, in subgroup analysis, this effect appears to be mostly driven by the impact in Asian patients.6

Until 2017, there were no approved agents for patients who did not respond positively to sorafenib. The results of the phase 3 RESORCE trial (OS for regorafenib: 10.6 versus 7.8 months in placebo; HR, 0.63; 95% CI, 0.50‐0.79) led to FDA approval of regorafenib as a second‐line therapy for patients with advanced HCC who progressed with sorafenib.7, 8 (Fig. 1) Notably, patients enrolled in the RESORCE trial were required to have tolerated sorafenib at a dose of at least 400 mg daily and maintain Child‐Pugh A cirrhosis and an ECOG status of 0 despite progression with sorafenib, which, when applied in clinical practice, is a highly selected population. Forty‐six percent of patients experienced grade ≥3 AEs in the trial; however, there was no meaningful difference in quality of life compared with placebo.7

Figure 1.

Treatment algorithm for advanced HCC.

Cabozantinib, another tyrosine kinase inhibitor, was shown to improve OS in patients who did not respond positively to first‐ and/or second‐line therapies in the phase 3 CELESTIAL trial. Cabozantinib also showed increased OS compared with placebo in patients with Child‐Pugh A with an ECOG of 0 to 1 (10.2 versus 8.0 months; HR, 0.76; 95% CI, 0.63‐0.92) and will be considered for approval as a second‐ or third‐line agent.9 Finally, the phase 3 REACH‐2 trial showed that ramucirumab improved OS as a second‐line agent in patients with preserved liver function and functional status who progressed or were intolerant to sorafenib with an alpha‐fetoprotein >400 ng/mL (OS, 8.5 versus 7.3 months; HR, 0.71; 95% CI, 0.53‐0.95) and thus will also be considered for approval in this setting in the coming months.10 (Table 1)

Table 1.

Studies for First‐ and Second‐Line Treatment for Advanced HCC

Author (Year)	Trial	Intervention	Control	Design	Patients (n)	Child‐Pugh Class A/B Score (%)	BCLC A/B/C (%)	Outcome	Outcome	P Value
First‐Line Treatment
Llovet3 (2008)	SHARP	Sorafenib	Placebo	RCT, phase 3	299 versus 303	95/5	0/17/83	Median OS	10.7 versus 7.9; HR, 0.69 (95% CI, 0.55‐0.87)	<0.001
Cheng15 (2009)	Asia‐Pacific	Sorafenib	Placebo	RCT, phase 3	150 versus 76	97/3	0/5/95	Median OS	6.5 versus 4.2; HR, 0.68 (95% CI, 0.50‐0.93)	0.014
Cheng16 (2013)		Sunitinib	Sorafenib	RCT, phase 3	530 versus 544	100/0	0/13/87	Median OS	7.9 versus 10.2; HR, 1.3 (95% CI, 1.13‐1.5)	0.99
Johnson17 (2013)	BRISK‐FL	Brivanib	Sorafenib	RCT, phase 3	577 versus 578	92/8	6/17/77	Median OS	9.5 versus 9.9; HR, 1.06 (95% CI, 0.93‐1.22)	0.37
Cainap18 (2015)		Linifanib	Sorafenib	RCT, phase 3	514 versus 521	95/5	0/15/85	Median OS	9.1 versus 9.8; HR, 1.046 (95% CI, 0.89‐1.22)	NS
Kudo6 (2018)	REFLECT	Lenvatinib	Sorafenib	RCT, phase 3	478 versus 476	99/1	0/21/79	Median OS	13.6 versus 12.3; HR, 0.92 (95% CI, 0.79‐1.06)	NS
In process	CheckMate 459	Nivolumab	Sorafenib	RCT, phase 3	726	100/0	Pending	Median OS	In process	In process
In process	HIMALAYA	Durvalumab +/− Tremelimumab	Sorafenib	RCT, phase 3	1200*	100/0	Pending	Median OS	In process	In process
Second‐Line Treatment (Sorafenib Failure or Intolerance)
Llovet3 (2013)	BRISK‐PS	Brivanib	Placebo	RCT, phase 3	263 versus 162	92/8	3/9/88	Median OS	9.4 versus 8.2 months; HR, 0.89 (95% CI, 0.69‐1.15)	0.104
Zhu19 (2014)	EVOLVE‐1	Everolimus	Placebo	RCT, phase 3	362 versus 184	98/2	0/14/86	Median OS	7.6 versus 7.3 months; HR, 0.93 (95% CI, 0.75‐1.15)	0.68
Zhu20 (2015)	REACH	Ramucirumab	Placebo	RCT, phase 3	283 versus 282	98/2	0/12/88	Median OS	9.2 versus 7.6 months; HR, 0.87 (95% CI, 0.72‐1.05)	0.14
Bruix7 (2017)†	RESORCE	Regorafenib	Placebo	RCT, phase 3	379 versus 194	98/2	0/14/86	Median OS	10.6 versus 7.8 months; HR, 0.63 (95% CI, 0.50‐0.79)	<0.0001
El‐Khoueiry11 (2017)	CheckMate‐040	Nivolumab	Sorafenib	Phase 1/2	262	98/2	BCLC‐C	ORR	15%‐20%	NA
Rimassa21 (2018)	Metiv‐HCC	Tivantinib	Placebo	RCT, phase 3	226 versus 114	95/5	7/12/81	Median OS	8.4 versus 9.1 months; HR, 0.97 (95% CI, 0.75‐1.25)	0.81
Abou‐Alfa9 (2018)	CELESTIAL	Cabozantinib	Placebo	RCT, phase 3	311 versus 466	100/0	BCLC‐C	Median OS	10.2 versus 8 months; HR, 0.76; 95% CI, 0.63‐0.92	0.005
Zhu10 (2018)	REACH‐2	Ramucirumab	Placebo	RCT, phase 3	197 versus 95	100/0	NR	Median OS	8.5 versus 7.3 months; HR 0.71; 95% CI, 0.53‐ 0.95	0.02
In process	KEYNOTE‐240	Pembrolizumab	Placebo	RCT, phase 3	408 (2:1)	A, B7	BCLC‐C	Median OS	In process	In process

^*Projected.

^†Sorafenib failure only.

Checkpoint Inhibitors

Checkpoint inhibitors, a form of immunotherapy, are increasingly being used in several solid malignancies, and they have been studied in HCC, with several trials scheduled to report results in the coming months. Nivolumab, a programmed death receptor‐1 (PD‐1) inhibitor, was studied in sorafenib‐naive and experienced patients in the phase 1/2 CheckMate‐040 trial with Child‐Pugh A liver function and an ECOG status of 0 to 1, which led to its FDA approval as a second‐line treatment for advanced HCC in 2017. The observed overall tumor response rate was 16%, with three or more AEs reported in 25% of patients.11 (Table 1) Nivolumab and other immunotherapy agents have the potential to induce immune‐mediated AEs, including autoimmune hepatitis, colitis, pneumonitis, and uveitis; however, the incidence of severe immune‐mediated reactions in the CheckMate‐040 was less than 5%. Nivolumab is not currently approved for use for patients with HCC in other countries, including throughout Europe, because of surrogate endpoint reporting in CheckMate‐040. A phase 3 clinical trial comparing nivolumab and sorafenib as first‐line therapy (CheckMate‐459) in patients with advanced HCC has enrolled and is pending full reporting (NCT02576509). Notably, the second‐line approval of nivolumab is contingent on the results of this trial showing superiority to sorafenib.

Pembrolizumab is another PD‐1 inhibitor that has recently attained conditional approval from the FDA for use as a second‐line treatment for advanced HCC (NCT02702401) (Table 2). Finally, a large phase 3 study of another checkpoint inhibitor, durvalumab, with and without tremelimumab (CTLA‐4 inhibitor), compared with sorafenib in the first‐line setting is currently recruiting (NCT03298451).

Table 2.

Locoregional Therapy for Advanced HCC

Author (Year)	Intervention	Control	Design	Patients (n)	Child‐Pugh Class B Score (%)	BCLC A/B/C (%)	MVI/EHS (%)	Survival, HR (95% CI)	P Value	Disease Control, HR (95% CI)	P Value
Yang22 (2012)	Cryotherapy + sorafenib	Sorafenib	RCT	52 versus 52	79/21	0/0/100	100/NA	Median OS 12.5 versus 8.6 months	0.01	TTP 9.5 versus 5.3 months	0.02
Luo23 (2011)	TACE	Conservative	Observational	84 versus 80	100/0	0/0/100	100/15	Median OS 7.1 versus 4.1 months; NA	<0.001	NA	NA
Pinter24 (2012)	TACE	Sorafenib	Observational	34 versus 63	59/41	0/0/100	32/41	Median OS 9.2 versus 7.4 months; NA	0.377	TTP 5.3 months	0.737
Kim25 (2015)	TACE	Sorafenib	Observational	295 versus 66	83/17 versus 66/34	0/0/100	100/11‐50	Median OS 5.9 versus 4.4 months; 0.57 (0.39‐0.83)	0.003	TTP 3.4 versus 1.8 months; 0.32 (0.19‐0.55)	<0.001
Yoon26 (2018)	TACE/RT	Sorafenib	Randomized, phase 2	45 versus 45	0	0/0/100	100/0	Median OS 12.8 versus 10.0 months; 0.61; (0.38‐0.98)	0.04	TTP 7.2 versus 2.7 months; 0.28 (0.17‐0.46)	<0.001
Kulik27 (2008)	SIRT (Y90)	Conservative	Phase 2	108	54/27	0/66/33	37/12	Median OS 16 months	NA	Partial response rate 42%‐70%	NA
D’Avola28 (2009)	SIRT (Y90)	Control	Observational	35 versus 43	94/6	3/51/46	46/NA	Median OS 16 versus 8 months;	<0.001	NA	NA
Mazzaferro29 (2013)	SIRT (Y90)	Conservative	Phase 2	52	82/17	0/66/34	100/25	Median OS 15 months	NA	TTP 11 months	NA
Kulik30 (2014)	SIRT (Y90) + sorafenib	SIRT (Y90)	RCT, phase 2	10 versus 10	70/30	60/10/30	7/NA	Survival rate 70% versus 72%	0.57	NA	NA
Edeline12 (2016)	SIRT (Y90)	Sorafenib	Observational	34 versus 107	82/18	0/0/100	53/NA	Median OS 26.2 versus 8.7 months; 0.4 (0.19‐0.82)	0.054	Response rate 78% versus 27%	0.003
Vilgrain13 (2017)*	SIRT (Y90)	Sorafenib	Randomized, phase 3	237 versus 222	84/16	4/28/68	60/0	Median OS 8 versus 9 months; 1.15 (0.94‐1.41)	0.18	PFS 4.1 versus 3.7 months; 1.03 (0.85‐1.25)	0.76
Chow14 (2018)†	SIRT (Y90)	Sorafenib	Randomized, phase 3	182 versus 178	NA	NA	NA	Median OS 8.8 versus 10 months; 1.1 (0.9‐1.4)	0.36	NA	NA
Song31 (2015)	HAIC (LFP)	Sorafenib	Observational	60 versus 50	84/16	0/0/100	100/33	Median OS 7.1 versus 5.5 months	0.011	TTP 3.2 versus 2.1 months	0.034
Moriguchi32 (2017)	HAIC (LFP) + sorafenib	Sorafenib	Observational	32 versus 14	100/0	0/0/100	100/22‐35	Median OS 10.3 versus 4 months	0.009	TTF 3.6 versus 1.2 months	0.002
Hatooka33 (2018)	HAIC ± sorafenib	Conservative	Phase 2	55	80/20	0/33/67	69/89	Median OS 19.9 months	NA	NA	NA
Bujold34 (2013)	SBRT	Conservative	Phase 1/2	102	100/0	1/34/65	55/12	NA	NA	Local control at 1 year 87%	NA
Nakazawa35 (2014)	3D CRT	Sorafenib	Observational	36 versus 28	59/41	0/0/100	19/19	Median OS 4.3 versus 5.9 months; NA	0.12	NA	NA

^*SARAH study.

^†SIRveNIB study.

Locoregional Therapy

Systemic therapy is recommended as the standard of care for advanced HCC in practice guidelines worldwide. However, liver‐directed locoregional therapies have been explored in advanced stage patients with portal vein tumor thrombus without extrahepatic disease. Observational cohort studies and early‐phase clinical trials with transarterial chemoembolization (TACE; with/without concurrent radiation therapy), hepatic arterial infusion chemotherapy (HAIC), and transarterial radioembolization with ⁹⁰Y‐loaded resin microspheres (selective internal radiation therapy [SIRT]) showed improvement in patient survival as compared with sorafenib.12 (Table 2) However, the recently completed phase 3 sorafenib versus radioembolization in advanced hepatocellular carcinoma (SARAH) and selective internal radiation therapy vesus sorafenib in locally advanced hepatocellular carcinoma (SIRveNIB) trials failed to demonstrate OS benefits of SIRT as compared with sorafenib in patients with Child‐Pugh A liver function and advanced HCC.13, 14 (Table 2) Even though these studies were negative, limitations related to the study designs (e.g., main portal vein invasion, limited ability to provide boosted radiation) may allow for further study of radioembolization in select patients with advanced stage HCC.

Surgery

Surgical resection can be an effective curative option in highly selected patients with portal vein tumor thrombus and preserved liver function. In observational studies from Asia, resection in patients with segmental or branch portal vein tumor thrombus show that 5‐year recurrence‐free survival rate can exceed 75%.36, 37 This approach warrants further study.

Best Supportive Care

Advanced HCC in the setting of decompensated cirrhosis and/or poor performance is terminal stage disease, and systemic therapies have not shown to be effective or safe in this population. Palliative care with the goal of symptom control should be discussed with all patients with advanced HCC and in particular in those patients without options for therapy.

Conclusions and Future Directions

Systemic therapies with sorafenib or lenvatinib are first‐line options in patients with advanced HCC, with several medications available or pending FDA review in the second and third line for first‐line failures (Fig. 1). We lack serum or tissue biomarkers to aid in therapy selection and lack adequate information on lines of therapy. Numerous clinical trials are due to report in the coming months that may expand treatment options for advanced HCC. In addition, there are planned or recently opened trials using combination therapy, including lenvatinib + pembrolizumab (NCT03713593) and cabozantinib + nivolumab. Finally, several adjuvant trials currently accruing are combining resection, ablation, or radiation therapy with immunotherapy to determine whether immunotherapy can enhance the effectiveness of these therapies or improve recurrence‐free survival. An open question remains whether immunotherapy can be safely used in patients who are being considered for liver transplantation and, if so, the optimal timing between receipt of immunotherapy and transplant.

Further understanding of comparative effectiveness of these therapies alone and possibly in combination will aid in developing evidence‐based treatment algorithms for patients with advanced HCC. Ultimately, precision medicine with development of biomarkers that can better direct therapy selection will be critical in the treatment of advanced HCC.