In de vorige richtlijnen was sorafenib de enig geregistreerde palliatieve behandeling voor het HCC. Nadien zijn er meerdere middelen in eerste en tweede lijn onderzocht en geregistreerd. Het doel van de module is de plaatsbepaling van deze middelen in de Nederlandse situatie te beschrijven.

First-line treatment

Conclusions

Overall survival (crucial)

Sorafenib versus placebo

Atezolizumab + Bevacizumab versus Sorafenib

Lenvatinib versus Sorafenib

Progression-free survival (crucial)

Sorafenib versus placebo

Atezolizumab + Bevacizumab versus Sorafenib

Lenvatinib versus Sorafenib

Tumour response rate (important)

Sorafenib versus placebo

Atezolizumab + Bevacizumab versus Sorafenib

Lenvatinib versus Sorafenib

Serious adverse events (important)

Sorafenib versus placebo

Atezolizumab + Bevacizumab versus Sorafenib

Lenvatinib versus Sorafenib

Quality of life (important)

Sorafenib versus placebo

Atezolizumab + Bevacizumab versus Sorafenib

Lenvatinib versus Sorafenib

Second-line treatment

Conclusions

Overall survival (crucial)

Regorafenib versus placebo

Progression-free survival (crucial)

Regorafenib versus placebo

Tumour response rate (important)

Regorafenib versus placebo

Serious adverse events (important)

Regorafenib versus placebo

Quality of life (important)

Regorafenib versus placebo

Third-line treatment

Conclusions

Overall survival (crucial), progression-free survival (crucial), tumour response rate (important), serious adverse events (important), quality of life (important)

First-line treatment

Four RCTs were selected that assessed the systemic therapy in the first-line setting.

Sorafenib versus placebo was studied in two RCTS (Llovet, 2008; Cheng, 2009). Lenvatinib versus sorafenib was studied in one RCT (Kudo, 2018). Atezolizumab + bevacizumab versus sorafenib was studied in 1 RCT (Finn, 2020) (see table 6.1).

Second-line treatment

1 RCT was selected that assessed the systemic therapy within the second-line setting.

This study assessed the efficacy and safety of regorafenib in patients with HCC who have progressed during sorafenib treatment (Bruix, 2017) (see table 6.2).

Third-line treatment

No studies reporting efficacy and safety of third-line treatment in patients with (unresectable) HCC were included in this literature review.

Description of studies

Subquestion 1 – Which first line systemic therapy is the most effective and safe for use in patients with (unresectable) HCC?

Sorafenib versus placebo

Llovet (2008) - SHARP described a phase III, double-bind, placebo-controlled trial, which was conducted in 21 countries (121 sites) in Europe, North America, South America, and Australasia. They evaluated the efficacy and safety of sorafenib in patients with advanced hepatocellular carcinoma who had not received previous systemic therapy. A total of 602 patients were randomized to receive sorafenib (n=299) at a dose of 400 mg twice daily or placebo (n=303). The mean (SD) age was 64.9 (11.2) years in the intervention group, compared with 66.3 (10.2) years in the control group. In the intervention group 260/299 (87%) were males, compared with 264/303 (87%) in the control group. All patients had a Child-Pugh class A score. The following relevant outcome measures were included: overall survival, time to symptomatic progression, objective response (partial/complete), adverse events. This RCT was included in the systematic review by Park (2021) and Oranratnachai (2021).

Cheng (2009) - described a phase III, double-bind, placebo-controlled trial, which was conducted in 23 sites within the Asia-Pacific region, in China, Taiwan, and South-Korea. They evaluated the efficacy and safety of sorafenib in patients with advanced hepatocellular carcinoma who had not received previous systemic therapy. A total of 226 patients were randomized to receive sorafenib (n=150) at a dose of 400 mg twice daily or placebo (n=76). The median (range) age was 51 (23-86) years in the intervention group, compared with 52 (25-79) years in the control group. In the intervention group 127/150 (84.7%) were males, compared with 66/76 (86.8%) in the control group. All patients had a Child-Pugh class A score. The following relevant outcome measures were included: overall survival, time to progression, objective response (partial/complete), adverse events, quality of life. This RCT was included in the systematic review by Park (2021) and Oranratnachai (2021).

Atezolizumab + Bevacizumab versus Sorafenib

Finn (2020) – IMbrave150 described a phase III, global, open-label randomized controlled trial, which was conducted at 111 sites in 17 countries. They evaluated the efficacy and safety of atezolizumab plus bevacizumab compared to sorafenib in patients with advanced hepatocellular carcinoma who had not received previous systemic therapy. A total of 573 patients were randomized to receive atezolizumab (IV, 1200 mg on day 1 of each 21-day cycle) plus bevacizumab (IV, 15 mg/kg on day 1 of each 21 day cycle (n=336), compared with sorafenib (orally administered 400 mg twice per day on days 1-21 of each 21-day cycle) (n=165). In the intervention group 277/336 (82%) were males, compared with 137/165 (83%) in the control group. All patients had a Child-Pugh class A score. The following relevant outcome measures were included: overall survival, progression-free survival, objective response (partial/complete), adverse events, quality of life.  This RCT was included in the systematic review by Park (2021) and Oranratnachai (2021). After the search date, a recent update became available which present updated data after 12 months of additional follow-up (Cheng, 2022). This study is added to the evidence table and data-synthesis.   

Lenvatinib versus Sorafenib

Kudo (2018) – REFLECT described a phase III, open-label, randomized controlled non-inferiority trial which was conducted at 154 sites in 20 countries throughout the Asia-Pacific, European, and North American region. They compared the overall survival in patients treated with lenvatinib versus sorafenib in patients with advanced hepatocellular carcinoma who had not received previous systemic therapy. A total of 954 patients were randomized to receive lenvatinib (12 mg/day for bodyweight ≥60 kg or 8 mg/day for bodyweight) (n=478) or sorafenib 400 mg twice-daily in 28-day cycles (n-476). In the intervention group 405/478 (85%) were males, compared with 401/476 (84%) in the control group. All patients had a Child-Pugh Class A score. The following relevant outcome measures were included: overall survival, progression-free survival, objective response (partial/complete), adverse events, quality of life.

Subquestion 2 – Which second line systemic therapy is the most effective and safe for use in patients with (unresectable) HCC?

Regorafenib versus placebo

Bruix (2017) - RESORCE described a phase III, double-bind, placebo controlled trial, which was conducted in 21 countries (152 sites) in North America, South America, Europe, Asia, and Australia. They evaluated the efficacy and safety of regorafenib in patients with advanced hepatocellular carcinoma who tolerated sorafenib (≥400 mg/day for ≥20 of last 28 days of treatment) and progressed on sorafenib. A total of 573 patients were randomized to receive regorafenib at a dose of 160 mg every day for 3 weeks of every 4-week cycle plus best supportive care (n=379) or placebo + best supportive care (n=194). The median (range) age was 64 (54-71) years in the intervention group, compared with 62 (55-68) years in the control group. In the intervention group 333/379 (88%) were males, compared with 171/194 (88%) in the control group. All patients had a Child-Pugh class A score. The following relevant outcome measures were included: overall survival, progression-fee survival, objective response (partial/complete), adverse events. This RCT was included in the systematic review by Park (2021) and Solimando (2022).

Subquestion 3 – Which third line systemic therapy is the most effective and safe for use in patients with (unresectable) HCC?

No studies reporting efficacy and safety of third-line treatment in patients with (unresectable) HCC were included in this literature review.

The Celestial study was not included as patients with both second- and third-line treatment were permitted without preplanned subpopulations efficacies.

First-line treatment

Overall survival (OS) (crucial)

Sorafenib versus placebo

Two studies (Llovet, 2008; Cheng, 2009) reported the median overall survival.

Llovet (2008) reported OS, which was measured from the date of randomization until the date of death from any cause. The median OS was 10.7 months in the intervention group, compared with 7.9 in de control group. This resulted in a HR of 0.69 (95% CI 0.55 to 0.87). This difference is considered clinically relevant.

Cheng (2009) reported OS, which measured from the date of randomization until the date of death from any cause. The median OS was 6.5 in the intervention group, compared with 4.2 in the control group. This resulted in a HR of 0.65 (95% CI 0.53 to 0.81). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the studies were RCTS. The level of evidence was downgraded by 2 levels, because of study limitations (risk of bias, -1 (see RoB assessment)) and because the optimal information was criteria is not met (imprecision, -1). Therefore, the level of evidence for the outcome ‘overall survival’ is considered low.

Atezolizumab + Bevacizumab versus Sorafenib

Cheng (2022) reported OS, which measured from the date of randomization until the date of death from any cause. The median OS was 19.2 in the intervention group, compared with 13.4 in the control group. This resulted in a HR of 0.66 (95% CI 0.53 to 0.81). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘overall survival’ is considered low.

Lenvatinib versus Sorafenib

Kudo (2018) reported OS, which measured from the date of randomization until the date of death from any cause. The median OS was 13.6 in the intervention group, compared with 12.3 in the control group. This resulted in a HR of 0.92 (95% CI 0.79 to 1.06). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘overall survival’ is considered low.

Progression-free survival (PFS) (crucial)

Sorafenib versus placebo

Cheng (2009) reported the time to progression. The median time to progression was 2.8 months in the intervention group, compared with 1.4 months in the control group. This resulted in a HR of 0.57 (95% CI 0.42 to 0.79). This difference is considered clinically relevant.

Llovet (2008) reported the time to symptomatic progression. The median time to symptomatic progression was 4.1 months in the intervention group, compared with 4.9 in the control group. This resulted in a HR of 0.58 (95% CI 0.45 to 0.74). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the studies were RCTS. The level of evidence was downgraded by 3 levels, because of study limitations (risk of bias, -1 (see RoB assessment)) and low number of patients (imprecision, -1), differences in outcome measures (indirectness, -1). Therefore, the level of evidence for the outcome ‘progression-free survival’ is considered very low.

Atezolizumab + Bevacizumab versus Sorafenib

Cheng (2022) reported PFS. The median PFS was 6.9 in the intervention group, compared with 4.3 in the control group. This resulted in a HR of 0.65 (95% CI 0.53 to 0.81). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘progression-free survival’ is considered low.

Lenvatinib versus Sorafenib

Kudo (2018) reported PFS. The median PFS was 7.3 in the intervention group, compared with 3.6 in the control group. This resulted in a HR of 0.64 (95% CI 0.55 to 0.75. This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘progression-free survival’ is considered low.

Tumour response rate (important)

Sorafenib versus placebo

Two studies (Llovet, 2008; Cheng, 2009) reported the response rate.

Llovet (2008) reported the tumour response rate, defined as the proportion of patients with a best response of complete response or partial response according to RECIST. There were no complete responses in either group. In the intervention group 7/299 (2.3%) achieved a partial response, compared with 2/303 (0.7%) in the control group. The RD was 1.7% (95% CI -0.26 to 3.6). This difference is not considered clinically relevant.

Cheng (2009) reported the tumour response, defined as the proportion of patients with a best response of complete response or partial response according to RECIST. There were no complete responses in either group. In the intervention group 5/150 (3.3%) achieved a partial response, compared with 1/76 (1.3%) in the control group. The RD was 2% (95% CI -1.8 to 5.9). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the studies were RCTS. The level of evidence was downgraded by 3 levels, because of study limitations (risk of bias, -1 (see RoB assessment)) and low number of events (imprecision, -2). Therefore, the level of evidence for the outcome ‘tumour response rate’ is considered very low.

Atezolizumab + Bevacizumab versus Sorafenib

Cheng (2022) reported the tumour response, defined as the proportion of patients with a best response of complete response or partial response according to RECIST 1.1. In the intervention group, 25/326 (7.7%) patients achieved a complete response and 72/326 (22.1%) a partial response. In the control group 1/159 (0.6%) patients achieved a complete response and 17/159 (10.7%) a partial response. The overall response rate was 97/326 (29.8%) in the intervention group compared to 18/159 (11.3%) in the control group. The RD was 18.4% (95% CI 11.4 to 25.4). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘tumour response rate’ is considered low.

Lenvatinib versus Sorafenib

Kudo (2018) reported the tumour response, defined as the proportion of patients with a best response of complete response or partial response according to a masked independent imaging review with a modified RECIST (mRECIST). In the intervention group, 10/478 (2.1%) patients achieved a complete response and 184/478 (38.5%) a partial response. In the control group 4/476 (0.8%) patients achieved a complete response and 55/476 (11.6%) a partial response. The overall response rate was 194/478 (40.6%) in the intervention group, compared to 59/476 (12.4%) in the control group. The RD was 28.2% (95% CI 22.9 to 33.5). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘tumour response rate’ is considered low.

Serious adverse events (important)

Sorafenib versus placebo

Two studies (Llovet, 2008; Cheng, 2009) reported the incidence of serious treatment-emergent adverse events.

Llovet (2008) reported the incidence of serious treatment-emergent adverse events.

In the intervention group was 52/297 (17.5%) patients experienced a treatment-emergent adverse event, compared with 54/302 (17.9%) in the control group. Diarrhea, weight loss, hand–foot skin reaction, and hypophosphatemia were more frequent in the sorafenib group (intervention). The RD was -0.4% (95% CI -6.5 to 5.8). This difference is not considered clinically relevant.

Cheng (2008) reported the incidence of serious treatment-emergent adverse events. In the intervention group, 71/149 (47.7%) patients experienced a treatment-emergent adverse event, compared with 34/75 (45.3%) in the control group. Hand-foot skin reactions, diarrhoea and fatigue were more frequent in the sorafenib group (intervention). The RD was 2.3% (95% CI -11.5 to 16.1). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the studies were RCTS. The level of evidence was downgraded by 3 levels, because of study limitations (risk of bias, -1 (see RoB assessment)) and low number of events (imprecision, -2). Therefore, the level of evidence for the outcome ‘serious adverse events’ is considered very low.

Atezolizumab + Bevacizumab versus Sorafenib

Cheng (2022) reported the incidence of serious adverse events and the incidence of serious treatment-emergent adverse events. In the intervention group, 76/329 (23.1%) patients experienced a treatment-emergent adverse event, compared with 25/156 (16.0%) in the control group. Proteinuria, hypertension and fatigue was most comment with atezolizumab plus bevacizumab (intervention), whereas palmar-plantar, erythrodysesthesia syndrome and diarrhea was most common in the sorafenib arm (control). The RD was 7.1% (95% CI -0.27 to 14.4). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘serious adverse events’ is considered low.

Lenvatinib versus Sorafenib

Kudo (2018) reported the incidence of serious treatment-emergent adverse events. In the intervention group, 205/476 (43.1%) patients experienced a treatment-emergent adverse event, compared with 144/475 (30.3%) in the control group. Hypertension, diarrhea, decreased appetite, and decreased weight was most common in the lenvatinib arm (intervention), whereas palmar-plantar erythrodysaesthesia, diarrhoea, hypertension, and decreased appetite was most common in the sorafenib arm (control). The RD was 12.8% (95% CI 6.7 to 18.8). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘serious adverse events’ is considered low.

Quality of life (important)

Sorafenib versus placebo

Quality of life was not reported.

Level of evidence of the literature

The level of evidence was not graded because no study reported quality of life.

Atezolizumab + Bevacizumab versus Sorafenib

Finn (2020) reported on the patient-reported outcomes, which were evaluated with the use of the European Organization for Research and Treatment of Cancer (EORTC) quality-of-life questionnaire for cancer (EORTC QLQ–C30). Finn (2020) reported the time to deterioration of quality of life, which was defined as decrease from baseline of 10 or more point on the EORTC QLQ–C30 maintained for two consecutive assessments or a decrease of 10 points or more in one assessment followed by death from any cause within 3 weeks. The median time to deterioration of patient-reported quality of life in the intervention group was 11.2 months (95% CI, 6.0 to not estimable), compared to 3.6 months (95% CI, 3.0 to 7.0) (HR 0.73, 95% CI 0.46 to 0.85) (favours atezolizumab + bevacizumab). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘quality of life’ is considered low.

Lenvatinib versus Sorafenib

Kudo (2018) reported the quality of life, which was assessed with the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire C30 (EORTC QLQ-C30)23,24 and the hepatocellular carcinoma-specific EORTC QLQ-HCC1825 health questionnaires. For between-group comparison, the summary score was not significantly different between the treatment arms (HR 0.87, 95% CI 0.75 to 1.01)(QLQ-C30 questionnaire). HCC18 questionnaire domains ranges between 0.79 (body image, favours lenvatinib) and 1.14 (pain, favours sorafenib). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘quality of life’ is considered low.

Second-line treatment

Overall survival (OS) (crucial)

Regorafenib versus placebo

Bruix (2017) reported the median OS, which was measured from the date of randomization until the date of death from any cause. The median OS was 10.6 months in the intervention group, compared with 7.6 in de control group. This resulted in a HR of 0.63 (95% CI 0.50 to 0.79). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘overall survival’ is considered low.

Progression-free survival (PFS) (crucial)

Regorafenib versus placebo

Bruix (2017) reported the median PFS. The median PFS was 3.4 months in the intervention group, compared with 1.5 in de control group. This resulted in a HR of 0.43 (95% CI 0.35 to 0.52). This difference is considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘progression-free survival’ is considered low.

Tumour response rate (important)

Regorafenib versus placebo

Bruix (2017) reported the objective response rate. In the intervention group, 2/379 (0.5%) patients achieved a complete response and 38/379 (10.0%) a partial response. In the control group 0/194 (0%) patients achieved a complete response and 8/194 (4.1%) a partial response. The overall response rate was 40/379 (10.6%) in the intervention group, compared to 8/194 (4.1%) in the control group. The RD was 6.4% (95% CI 2.3 to 10.6). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘tumour response rate’ is considered low.

Serious adverse events (important)

Regorafenib versus placebo

Bruix (2017) reported the incidence of serious adverse events.

In the intervention group was 166/397 (41.8%) patients experienced a serious adverse event, compared with 90/194 (46.3%) in the control group. Diarrhea, weight loss, hand–foot skin reaction, and hypophosphatemia were more frequent in the sorafenib group (intervention). Hypertension, hand-foot skin reaction, fatigue, and diarrhea were more frequent in the regorafenib group (intervention). The RD was -2.6% (95% CI -11.2 to 6.0). This difference is not considered clinically relevant.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘serious adverse events’ is considered low.

Quality of life

Regorafenib versus placebo

Quality of life was assessed with the Functional Assessment of Cancer Therapy (FACT-G), FACT-Hepatobiliary (FACT-Hep), EQ-5D and EQ-VAS. No clinically meaningful differences were noted between the regorafenib and placebo groups in HRQoL.

Level of evidence of the literature

The level of evidence started as high, because the study was an RCT. The level of evidence was downgraded by 2 levels, because the optimal information size criteria is not met (imprecision, -2). Therefore, the level of evidence for the outcome ‘quality of life’ is considered low.

A systematic review of the literature was performed to answer the following question:

Which treatment is to be advised for patients with (unresectable) HCC?

PICO

P: Non-locally-treatable patients with a diagnosis of hepatocellular carcinoma;

I: Systemic therapy;

C: Other systemic therapy or placebo;

O: Overall survival, progression-free survival, response rate, adverse events, quality of life.

Relevant outcome measures

The guideline development group considered overall survival and progression-free survival as a crucial outcome measure for decision making; and response rate, complications/adverse events and quality of life as important outcome measures for decision making.

The guideline development group defined the outcome measures as follows:

Clinically relevant difference

The guideline development group defined a minimal clinically relevant difference at a minimum ofsof:

• Overall survival: >12 weeks or HR <0.7.
• Progression-free survival: >12 weeks or HR <0.7.

And, in case of absence of a clinically relevant difference in overall survival or progression-free survival:

• Response rate: >20% difference.
• Quality of life: A minimal clinically important difference of 10 points on the quality of life instrument EORTC QLQ-C30 or a difference of a similar magnitude on other quality of life instruments.
• Adverse events: <25% difference.

Data-synthesis

Results from RCTs studies were described and synthesized (preferably by meta-analysis) separately.

Search and select (Methods)

The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms until April 15^th, 2022. The detailed search strategy is depicted under the tab Methods.

The systematic literature search resulted in 1115 hits. Studies were selected based on the following criteria:

• included patients HCC, ineligible for surgical intervention;
• compared systemic therapy;
• reported at least one of the outcomes of interest;
• the study design is a systematic review (SR) (preferably of randomized controlled trials; RCTs), or RCT;
• written in English language.

Based on title and abstract screening, 14 studies were initially selected. After reading the full text and thorough assessment of the studies, 11 studies were excluded (see table with reasons for exclusion under the tab Methods), and 3 systematic reviews were included.

Results

A total of three studies were included in the analysis of the literature. The important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables.

Not all systemic therapies are included in the literature summary. The Committee BOM (cieBOM) makes recommendations about the procedure on the basis of the evidence relating to its efficacy and safety. Therefore, we will answer the question which of the available therapies in the Netherlands is the most effective and safe in patients with (unresectable) HCC in the first-line, second-line and third-line. An overview of all the included RCTs in the systematic reviews and the selected RCTs can be found in table 6.1 and table 6. 2.

The first line available treatments in the Netherlands are:

• Sorafenib.
• Atezolizumab + Bevacizumab.
• Lenvatinib.

The second- and further line available treatments in the Netherlands are:

• Sorafenib (add-on).
• Regorafenib.

Currently, no drugs of proven efficacy in the third-line settings are available in the Netherlands.