A case of patient with advanced hepatocellular carcinoma - maximizing clinical benefits through timely switching away from failing regimen


Dr. Chiang, Chi-Leung

Clinical Assistant Professor,
Department of Clinical Oncology,
Li Ka Shing Faculty of Medicine,
The University of Hong Kong

Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide and accounts for approximately 5.7% of all new cancer cases annually.1 In Hong Kong, HCC is the fourth and seventh most common cancer in men and women, respectively, and is the third leading cause of cancer-related death.2 In advanced HCC cases such as those with extrahepatic spread or portal vein invasion, systemic targeted treatment with sorafenib is a standard of care option in Hong Kong and was shown to prolong survival in patients who had progressed after surgery or locoregional therapy.2,3 With continuous development in the past decade, other systemic treatment options such as lenvatinib, regorafenib, and atezolizumab plus bevacizumab (atezo/bevac) had become available for HCC treatment.4,5 In the current case shared by Dr. Chiang, Chi-Leung, clinical assistant professor from the Department of Clinical Oncology, the University of Hong Kong, the patient was switched from one systemic treatment to another at early sign of treatment failure as guided by serum biomarker elevation and/or radiological imaging. Given the availability of a good number of systemic options, rapid switching from failing regimen may be a feasible strategy, allowing HCC patients to receive multiple lines of optimized care that would have impact on survival as well as quality of life (QoL).

Case Review

The 52-year-old male patient, having no family and medical history of cancer, was first presented to the hospital with right epigastric pain in February 2019 in Hong Kong. Computer tomography (CT) workup identified HCC with main tumor of 10cm at liver segment IVa with right anterior portal vein invasion (Vp2) and another smaller 2.1cm tumor at liver segment V. Scheduled resection was aborted due to rapid disease progression within a month with enlargement of main tumor to 18cm and emergence of smaller tumors in segment IVa and V.

The patient was referred to ablative stereotactic body radiation therapy (SBRT) followed by immunotherapy (IO) in March 2019. After 35Gy in 5 fractions, alpha-fetoprotein (AFP) dropped >80% from 2,757 to 471ng/mL in 2 weeks and nivolumab was initiated as IO in April 2019 to harness the synergistic benefit with SBRT as studies have suggested that SBRT may prime the tumor microenvironment in favor of IO.6 Unexpectedly, the patient was hyper-progressive to IO as reflected by biochemical (AFP rebounded to 2,366ng/mL within 1 month) and radiological indicators in 1.5 months. Imaging revealed multiple new hypervascular lesions appeared outfield in both liver lobes with huge 17cm conglomerate mass in left liver as well as several nodal/pulmonary metastases in the thorax despite shrinkage of irradiated main tumor at right liver. These were also associated with intensified epigastric pain and plummeted liver function (bilirubin 54μmol/L, albumin 28g/L, Child-Pugh B8). IO was discontinued immediately after imaging.

Two weeks later in late May 2019, the patient was prescribed lenvatinib as multikinase inhibitor (MKI) for tumor control after recovery of liver function. AFP, initially dropped from 5,612 to 3,812ng/mL, rebounded to 8,648ng/mL within 1 month with multiple outfield progressions in the liver with thrombosed left portal vein (Vp3) and the thorax. Drug-related liver toxicity was noted at week 3 that was ameliorated on dose-reduction. In early July 2019, the patient was switched to regorafenib, another MKI, at reduced dosage of 80mg/day considering impaired liver function (total bilirubin: 36μmol/L, albumin 31g/L, Child-Pugh B7). The patient had remarkably treatment response with >99% AFP drop in 6 weeks (from 8,648 via 326 to 65ng/mL) followed by normalization in month 4 (Figure 1). Target tumors at liver and thorax (nodal and pulmonary) showed interval reduction in size with liver masses having significant reduction in arterial enhancement on CT in week 7 and month 4 after regorafenib treatment that was considered very good partial response (Figure 2). Patient had liver function improved (bilirubin normalized to 12μmol/L, albumin 38g/L, Child-Pugh A5) and painful epigastric symptoms relieved. He tolerated regorafenib very well probably due to the reduced dosing all along. He felt energized on regorafenib and went for leisure travel with his wife overseas. The clinical, biochemical, and radiological improvement lasted for 6 months until December 2019 where most target tumors started to regrow with AFP increased to 351 from a nadir of 9ng/mL.

Sorafenib was attempted in January 2020 as another MKI but did not stop the rising trend of AFP. Treatment was quickly adapted to atezo/bevac in February 2020 after discussion with the patient on risk/benefit of this boosted IO combo due to prior history of hyper-progression disease (HPD). Closer monitoring on biochemistry and immuno-related symptoms were in place. A significant 90% drop in AFP from 3,424 to 122ng/mL was observed in 12 weeks with pain relieved and general condition improved. By March 2020, HCC with lung and lymph node metastases were static radiologically. Side effects were manageable with moderately elevated liver enzymes and mild decrease in platelets. As of July 2020, atezo/bevac is ongoing for 4 months and a slight rebound in AFP was observed.

If disease further progresses, cabozantinib or double IO are likely the next options. As the patient had previous HPD to anti-PD1 therapy, cabozantinib will be the preferred option and double IO will be reserved after cabozantinib treatment fails.


As most advanced HCC patients do not survive beyond second-line treatment, this patient case is particularly impressive given that he has survived for 15 months+ through 5 lines of non-surgical treatment with acceptable QoL without hospitalization despite an initial diagnosis of unresectable HCC of 18cm, portal vein invasion plus numerous smaller lesions in the liver. Bear in mind his treatment journey has not been so straight forward, and the followings deserve discussion.

The initial SBRT/nivolumab combo, despite both being effective on its own and the scientific rationale of priming the immune system through radiation for anti-PD1 immunotherapy to act on, did not achieve the intended goal of local control for inoperable HCC.6-11 While SBRT has significantly shrunk the original huge tumor from 18 to 8cm in 3 months, nivolumab induced HPD has led to AFP surge, numerous new lesions formed in the non-irradiated liver, lung metastasis, intensified epigastric pain, and plummeted liver function within 1 month. HPD could be fatal and is likely to deprive the patient a chance for next step therapy. Recent retrospective Korean cohort analysis by Kim et al. revealed that for the first time in HCC, HPD on nivolumab was associated with dismal median overall survival of <2 months and happened in 12.7% (24/189) of patients.12 Notably, the authors defined HPD rather stringently as 4 times (higher than the typical 2 times in other cancers) increase in both tumor growth kinetics (TGK) and tumor growth rate (TGR) before and after immunotherapy plus >40% absolute increase of TGR, which may have resulted in an estimate that is on the conservative side. Common clinicopathological factors that implicated HPD in other cancers, including age (>65yrs), albumin (<3.5g/L), elevated lactate dehydrogenase (LDH), and multiple metastases (>2 sites), were not associated with HPD in HCC except high baseline neutrophil to lymphocyte ratio (NLR >4.125; sensitivity 87.5%, specificity 74.5%). Whether high NLR could readily predict HPD in HCC is pending for verification but may not be used as an exclusion factor for IO. Instead, NLR may be discussed with the patient as a part of the risk/benefit analysis that help alert both patient and physician to closely monitor the initial treatment for the sake of early detection of abnormalities and remedies. In the current case, baseline NLR was not checked but the unexpected surge of AFP and associated symptoms after putting in IO have alerted for closer monitoring and earlier imaging for confirmation. Fortunately, the patient was quickly shifted to another systemic therapy and has responded well.

Another interesting observation was the regain of tumor control by regorafenib after unexpected failure of nivolumab and perhaps lenvatinib. Rapid tumor shrinkage at all sites to very good partial response lasting for half a year was unexpected given the tumor static nature of regorafenib seen in the clinical trial.13 Additionally, plummeted liver function was also normalized for this patient as well. Currently, the patient selection criteria for regorafenib and other systemic therapies in HCC is unclear as predictive biomarkers are still lacking. However, rapid switching from failing regimen has been the key to this patient’s survival by exposing the patient to other active agents that may work before liver function is exhausted. While bringing old concept new, tracking AFP changes helps prognostication in HCC for those with elevated AFP. Early decline of AFP (at week 6/8), especially sharp decline, is likely suggestive of better treatment outcomes as seen in the current case (during treatment with regorafenib or atzeo/bevac) as well as in a number of post-hoc analyses of phase 3 HCC clinical trials and real-life studies.14-19 That said, the threshold for surrogacy of AFP remains to be established, and imaging could always be used on top of AFP dynamics to ascertain disease status, especially in cases that have a rising trend. Good use of AFP dynamics ± imaging often permits earlier pickup of “silent” progression and timely switch to next therapy can give the patient another chance.

Now, we are not in the shortage of active HCC therapies. The next challenge is how to optimally place them in a row for HCC patients, a luxurious trouble that could hardly have during the time of only 1 active agent. To this end, personalization based on risk/benefit evaluation of the patient shall be key.


With the number of available systemic treatment options increased, switching to another treatment option as early as first sign of treatment resistance is a feasible strategy for treating advanced HCC patient. However, the question in regimen sequencing remains a challenge in the treatment landscape of HCC.

Get access to our exclusive articles.