Given the generally dismal outcomes in HCC, patients with HCC who are not candidates for resection or transplantation should be considered for clinical trials whenever possible. There is a growing understanding of the signaling pathways (eg, Ras/Raf/MAPK, WNT-β-catenin, EGFR, insulin-like growth factor receptor, AKT-mTOR, Notch, and Hedgehog) that drive hepatic carcinogenesis, and their components represent potential molecular targets for therapy in HCC (Table 3).9 Somatic genomic profiling using multiplex assays may be a way to enrich clinical care by detecting actionable mutations, though the percentage of patients with HCC with actionable mutations in a recent prospective clinical sequencing study was less than 10%.10 Creating tools to match patients with actionable mutations with available clinical trials is an area of ongoing investigation, and the number of “actionable” mutations is expected to increase as basic and clinical sciences increasingly focus on precision oncology.11 For all of our patients with HCC who need systemic therapy, we strongly advocate for participation in clinical trials.
(To view a larger version of Table 3, click here.)
Cytotoxic chemotherapy plays a minor role in the treatment of advanced HCC. Doxorubin was long considered the most effective chemotherapeutic agent in the treatment of advanced HCC, but recent evidence refutes any significant benefit. Alliance/CALGB 80802 compared sorafenib alone to sorafenib in combination with doxorubicin in 346 patients with advanced HCC. Patients treated with combination therapy experienced more toxicity and inferior OS (9.3 months vs 10.5 months for sorafenib) or progression-free survival (PFS).12 A randomized Phase III study of doxorubicin versus cisplatin/interferon alpha-2b/doxorubicin/fluorouracil (PIAF) showed modest overall improvements in overall response rate and median survival with PIAF. However, the results were not statistically significant, and treatment-related toxicity was observed in patients receiving PIAF.13 Somewhat more promising data support the use of FOLFOX (infusional fluorouracil, leucovorin, and oxaliplatin) and GEMOX (gemcitabine-oxaliplatin). A Phase III trial involving 371 Asian patients with advanced HCC comparing modified FOLFOX and doxorubicin (50 mg/m2 every 3 weeks) showed improved PFS for FOLFOX (2.93 months vs 1.77 months; HR=0.62; 95% CI=0.49–0.79; P<0.001). The primary endpoint of difference in OS was not met, but there was a trend toward improvement with FOLFOX (HR=0.79; 95% CI=0.63–0.99; P=0.04). Proportions of grade 3 or 4 adverse events were similar between treatment groups.14 It is unclear whether these findings in patients with predominantly hepatitis B virus (HBV)-associated HCC can be extrapolated to American patients with predominantly HCV- and NAFLD-associated HCC.
A number of small, single arm trials suggest that GEMOX is a good option for advanced HCC. In a study, 21 patients with advanced HCC demonstrated an overall response rate of 19%(95% CI=13%–26%), median PFS of 5 months, and median OS of 12 months.15 In another study, in 32 patients with previously untreated advanced HCC, GEMOX produced a response rate of 18% (95% CI=8%–34%), median PFS of 6.3 months, and OS of 11.5 months. For unclear reasons, treatment was significantly more effective in patients with nonalcoholic rather than alcoholic cirrhosis.16 In both trials, GEMOX was reasonably well tolerated.
A meta-analysis of 13 studies using first-line oxaliplatin-based chemotherapy demonstrated a pooled response rate of 16.8%. The median PFS and OS were found to be 4.2 and 9.3 months, respectively, and the 1-year OS was 37%.17 As the majority of data on chemotherapy in HCC is from the first-line setting, there is substantial uncertainty regarding the tolerability of such an approach following sorafenib failure. However, for medically fit patients who have progressed on sorafenib and regorafenib and who are unable to participate in a clinical trial, we recommend the use of FOLFOX or GEMOX.
Several hormonal agents studied in HCC, including tamoxifen, megestrol, octreotide, and lanreotide, have failed to show meaningful clinical benefit in HCC. Due to the presence of estrogen receptors on one-third of HCCs, estrogen receptor blockade with tamoxifen was hypothesized to have potential benefit in HCC. Four randomized controlled trials (RCTs) and a Cochrane systematic review of tamoxifen in HCC failed to show improvement with respect to survival or functional status.18–22 Data are conflicting regarding the use of megestrol in HCC. A small RCT comparing megestrol versus supportive care alone showed improved median OS with megestrol (18 months, 95% CI=13.47–22.53 months) versus untreated patients (7 months, 95% CI=3.01–10.99, P=0.0090).23 However, a larger randomized placebo-controlled trial involving 204 patients with treatment-naive advanced HCC showed no improvement in OS, and in fact showed worse outcomes amongst patients with poor performance status or Child Pugh class B or C who received megestrol.24 Somatostatin analogs, such as octreotide and lanreotide, have been studied in HCC and have shown some survival benefit in patients with positive octreotide scans25 but have failed to show consistent improvement with respect to PFS or OS.26–30 The balance of evidence suggests that hormone therapies are not effective against HCC and should not be considered part of clinical practice.
Molecularly targeted therapies
Since sorafenib’s approval, there have been multiple trials comparing first-line sorafenib to a novel targeted agent or combined with another agent that have failed to show any clinically relevant benefit over sorafenib.12,31–33 A variety of drugs have also failed to show benefit in the second-line setting.34–37The two key exceptions are lenvantinib and regorafenib, both multi-target TKIs that have recently met their primary endpoints in first- and second-line therapy, respectively.
Regorafenib is an orally active nonspecific TKI that works by inhibiting several pro-neoplastic pathways, including those involved in angiogenesis, growth, and metastasis. In April 2017, the FDA approved regorafenib for second-line use in patients with HCC who were previously treated with sorafenib on the basis of the RESORCE trial, a randomized double-blind placebo controlled trial of the efficacy of regorafenib in advanced, refractory HCC.38,39 This study of 573 patients with ECOG PS 0-1 and Child Pugh A cirrhosis whose tumors had progressed after receiving sorafenib demonstrated a median OS of 10.6 months with regorafenib versus 7.8 months with placebo (HR=0.63; 95% CI=0.50–0.79; one-sided P<0.0001). The median PFS was found to be 3.1 months with regorafenib versus 1.5 months with placebo. The profile of clinically relevant adverse events were as expected from regorafenib including hypertension (15% in regorafenib group versus 5% in placebo group), hand–foot syndrome (13% versus 1%), fatigue (9% versus 5%), and diarrhea (3% versus 0%). The recommended dose of regorafenib is 160 mg once daily on days 1–21 of a 28-day cycle.39 This is the first new FDA-approved drug for HCC in more than a decade and represents the only FDA-approved drug for second-line treatment of HCC.
Lenvatinib, another oral multikinase inhibitor that targets VEGFR, FGFR, PDGFR-beta, RET, and KIT, demonstrated similar activity as sorafenib as a first-line treatment for advanced HCC.40 In the open-label REFLECT study, 954 patients with untreated HCC were randomized to receive lenvatinib or sorafenib as initial therapy. The trial met its primary endpoint, showing lenvatinib to be noninferior to sorafenib (median OS 13.6 vs 12.3 months, HR=0.92, 95% CI=0.79–1.06). Median PFS was better in lenvatinib-treated patients (7.4 months vs 3.7 months, HR=0.66, 95% CI=0.57–0.77), as was the disease control rate of 71.7% versus 60.5%. Lenvatinib did not, however, have a better adverse effect profile. Patients reported similar declines in health-related quality of life after treatment initiation in both groups and similar rates of severe toxicities.40