Spontaneous regression of metastatic hepatocellular carcinoma following 3 weeks of lenvatinib

  1. Lauren Curry 1,
  2. Warda Limaye 2 and
  3. Ravi Ramjeesingh 1 , 3
  1. 1 Department of Medicine, Nova Scotia Health Authority, Halifax, Nova Scotia, Canada
  2. 2 Department of Diagnostic Radiology, Nova Scotia Health Authority, Halifax, Nova Scotia, Canada
  3. 3 Division of Medical Oncology, Nova Scotia Health Authority, Halifax, Nova Scotia, Canada
  1. Correspondence to Dr Ravi Ramjeesingh; ravi.ramjeesingh@nshealth.ca

Publication history

Accepted:19 Jan 2022
First published:09 Feb 2022
Online issue publication:09 Feb 2022

Case reports

Case reports are not necessarily evidence-based in the same way that the other content on BMJ Best Practice is. They should not be relied on to guide clinical practice. Please check the date of publication.

Abstract

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality and is associated with a poor prognosis. Rarely, there is spontaneous regression of the tumour. We present a case of a middle-aged male with presumed metastatic HCC who experienced enduring regression following 3 weeks of lenvatinib, which was discontinued due to side effects. While this could represent an unusually successful response to therapy, spontaneous tumour regression or an alternative diagnosis should be considered. We discuss possible mechanisms that might explain this unusual case and advocate for tissue confirmation in select cases, where there is diagnostic doubt or when the disease pattern does not clearly follow the recognised natural history. Therefore, if regression occurs—whether spontaneous or in response to treatment—it can be better understood and subsequent therapies recommended.

Background

Hepatocellular carcinoma (HCC) is the most common histological type of primary liver cancer and is a major cause of cancer-associated mortality worldwide.1 An estimated 70%–80% of cases develop in patients with liver cirrhosis, limiting the feasibility of surgical resection.2 A patient’s hepatic reserve, as indicated by the Child-Turcotte-Pugh classification, often dictates therapeutic options, including resection, radiation, radiofrequency ablation, transarterial chemoembolisation (TACE) and systemic treatment including molecularly targeted and immunological agents.

Lenvatinib, an oral multityrosine kinase receptor inhibitor of vascular endothelial growth factor receptors 1–3, fibroblast growth factor receptor 1–4, platelet-derived growth factor receptor alpha, RET and KIT oncogenes, is considered a first-line therapy for cases of advanced HCC.3 4 When targeted therapy is indicated, lenvatinib is often preferred over sorafenib due to better tolerability and the higher objective response rates and longer time to tumour progression in the REFLECT trial.4 Despite the availability of treatment options, the prognosis of patients with advanced HCC remains poor, with a 5-year survival rate of only 18%.5

Cole and Everson first defined spontaneous tumour regression (SR) in 1956, as a partial or complete involution of a malignant tumour in the absence of specific therapy or in the presence of therapy considered inadequate to exert significant benefit on neoplastic disease.6 It is a rare phenomenon, with an estimated incidence of one of every 60 000–100 000 cases.6 A meta-analysis of 10 phase III HCC treatment trials over 30 years reported a spontaneous partial regression rate of 0.4%.7 Early literature identified malignant melanoma, neuroblastoma and renal cancers as more commonly associated with SR.8 More recently, however, lung and liver cancers have also been associated with increased rates of SR. There are several case reports of HCC SR in the literature; however, the aetiology remains unclear.

Case presentation

A middle-aged male presented to the local emergency department with chest pain. A CT aortic dissection protocol incidentally revealed multiple lesions suspicious for metastatic cancer. Medical history included liver cirrhosis secondary to alcohol and hepatitis C (sustained viral response following treatment), chronic obstructive pulmonary disease, peripheral vascular disease, coronary artery disease and porphyria cutanea tarda. He was an active smoker and had quit alcohol after years of misuse. Two years prior, there was no evidence of HCC on abdominal ultrasound. Prior to this presentation, his alpha fetoprotein (AFP) levels had been slowly rising (figure 1). The Child-Pugh score was 6 /A.

Figure 1

Our patient’s AFP over time. Normal range 10–20 ng/mL. AFP, alpha fetoprotein.

Investigations

Initial CT imaging demonstrated an irregular liver contour with multifocal hyperenhancing hepatic lesions in the left and right lobes, with large heterogeneously enhancing adrenal masses (R: 7.7×6.5 cm, L: 6.3×5 cm) (figure 2). No definite primary was seen. AFP at this time was 1401 ng/mL (10–20 ng/mL) and peaked at 3619 ng/mL (figure 1). Two subsequent liver biopsies were unfortunately non-diagnostic and revealed only necrotic tissue with chronic inflammation and background (grade 2/4, stage 4/4) hepatitis C virus-related cirrhosis. A dedicated liver MRI showed few enhancing lesions in segments 4 and 8 (Li-RADS 3; intermediate HCC risk) and numerous non-enhancing and centrally necrotic hepatic lesions (largest 16 mm) not meeting criteria for HCC, in addition to necrotic bilateral adrenal masses, stable lytic bone lesions and background features of cirrhosis. A subsequent adrenal gland biopsy was unfortunately again non-diagnostic and demonstrated only necrotic tissue.

Figure 2

Fifty-year-old man with new hepatic lesions and bilateral adrenal masses initially identified on CT exam with intravenous contrast (A). Follow-up imaging with MRI was performed (B–D). (B) T1-weighted imaging with intravenous contrast further characterises the hepatic lesions, including a 1.3 cm enhancing Li-RADS 3 lesion in hepatic segment 4 and the necrotic appearing adrenal masses. (C) Interval decrease in the size of the hepatic segment 4 lesion, measuring 9 mm. (D) Slight increase in the hepatic segment 4 lesion, which does not meet criteria for interval growth with decreased size of the adrenal masses bilaterally.

Differential diagnosis

The radiographic findings were highly suspicious for malignancy of unknown primary, affecting the liver and adrenal glands. The multifocality made benign processes less likely (eg, adenoma and haemangioma). An inflammatory pseudotumour was possible but felt less likely given no clear inciting inflammatory event and evidence of necrosis on biopsy. Given his history of cirrhosis and elevated AFP, HCC could not be excluded and was felt to be the most likely diagnosis. His case was reviewed at the local hepatopancreaobiliary tumour boards, and it was agreed on that the patient had metastatic HCC and was not a candidate for curative intent treatment.

Treatment

Due to the metastatic and multifocal nature of disease, he was determined not to be a candidate for transplant, resection or local treatment. He underwent radiation (800 Gy, one fraction) to bony lesions of the sacrum and pelvis, which resulted in symptomatic relief. Shortly thereafter, palliative lenvatinib was started at 12 mg orally daily, prior to the reported MRI. The patient experienced nausea, vomiting, fatigue and depression and therefore stopped taking lenvatinib after 3 weeks.

Outcome and follow-up

Interestingly, AFP declined to 114 ng/mL after treatment and had normalised 2 months later (figure 1). A repeat MRI showed decreased disease burden, including the Li-RADS 3 enhancing lesions more concerning for HCC, and further reduction of the bilateral adrenal masses and volume of ascites (figure 2). Normal AFP values persist now, 2 years after the initial diagnosis. Imaging completed just prior to the submission of this manuscript demonstrated an ongoing reduction in decrease burden. However, the segment 3 lesion continues to be Li-RADS 3 (figure 2). Today, the patient continues off treatment and is clinically stable with no evidence of relapse or rising tumour markers. Future local treatment is being considered.

Discussion

Spontaneous tumour regression is a rare phenomenon in the field of oncology. More than 90 cases of HCC SR have been reported in the literature; the first partial SR was reported in 1972 and complete SR in 1982.9–11 HCC SR with distant metastases is a far rarer event.9 In most cases, there is underlying virus-induced hepatic disease.12 In a recent review, the duration between diagnosis and SR is short, with an average of 4 months.13 The underlying mechanisms are unclear but can be broadly categorised as tissue hypoxia and systemic immunological response.14 Regression has been observed in cases of tissue hypoxia following arterial thrombosis,15 rapid tumour expansion that outpaces neovascularisation16 17 and systemic hypoperfusion (eg, life threatening bleed)18—there were no such scenarios in this case. Several authors have attributed regression of primary and metastatic HCC lesions to the abscopal effect of radiation therapy, in which localised radiation treatment is associated with cancer regression at a site distant from that irradiated.19–21 Postradiation regression has been associated with an increase in circulating Tumour Necrosis Factor (TNF)-alpha and other inflammatory markers, lending support to an immune-mediated mechanism of regression, potentially by ‘priming’ the immune system resulting in antitumour effects outside of irradiated areas.9 This phenomenon could be relevant where our patient received palliative radiation to his sacrum and pelvis prior to SR. Most of all, we question the potential immune response induced by a short treatment of lenvatinib on tumour carcinogenesis in our patient.

Phase II and III lenvatinib trials reported a partial response (PR) in 37% and 38% of participants, respectively, based on masked independent imaging review using the modified Response Evaluation Criteria in Solid Tumors (mRECIST).3 4 However, the timeline of observed response is unclear from the study methods and likely required more than 3 weeks of lenvatinib, as taken by our patient. A more recent study assessed the early therapeutic effects of lenvatinib on unresectable HCC using contrast-enhanced CT images obtained between 4 and 8 weeks after starting lenvatinib, also according to mRECIST.22 Among 54 patients, they observed a PR in 44.4%, and early therapeutic efficacies differed depending on the extent of HCC staging and baseline AFP.23 Similar results were obtained in real-world practice; Hiraoka et al reported a complete or PR achieved in 40.7% of patients with unresectable HCC at 4 weeks after the initiation of lenvatinib.24

Our patient’s diagnosis was presumed metastatic HCC given his history of cirrhosis, elevated AFP and radiographic findings. However, all biopsies were non-diagnostic of tumour cells, revealing only necrotic tissue. An alternative diagnosis is therefore possible. In the literature, many cases of spontaneous HCC regression lacked a tissue biopsy prior to the observed regression, with the initial diagnosis made similarly to our case (ie, radiographically).9 19 24–28 What is perhaps more interesting are the results of pathology following cases of SR. Most observed some viable cancer cells with necrosis and inflammation,9 12 24–26 sometimes with a thick fibrous capsule or thrombi.12 24 25 However, others observed only necrotic tissue without evidence of cancer cells, as in our case.25 In these scenarios, the authors also questioned if they had the correct diagnosis.25 27

We decided to explore this further with a Novanet search of SR cases written in English between 1990 and 2021 including the title words ‘hepatocellular carcinoma’ and ‘regression’ and any field containing ‘spontaneous regression’. All resulting articles were reviewed and assessed for inclusion. Fifty-one articles containing 61 cases were included, detailed in table 1. We identified that SR occurred following recurrence in 39% and 33% in cases of distant metastases. Treatments prior to SR were variable. SR was established radiographically in most cases; 57% noted PR and 43% complete response. HCC was confirmed with a tissue diagnosics in 61% of cases, either pre-SR or post-SR. Examination of surgically resected specimens following SR occurred in 33% of cases, with 60% showing residual HCC. To assess for SR duration, we addressed the variability in end points (ie, study end, recurrence, definitive treatment and death) separately. For all end points, the median and mean SR durations were 23 and 20 months, respectively. For studies ending with sustained SR, the median and mean durations were 24 and 30 months, respectively. For all other end points (ie, recurrence, treatment and death), the median and mean SR durations were each 12 months. We observed a 26% recurrence rate after SR. Another review of 28 cases estimated a relapse rate of 25%, which was significantly less likely to occur when additional therapies such as surgical treatment, radiation and TACE occurred. Taken together, there is great variability among cases of HCC SR in terms of preregression treatment, diagnosis and postregression pathology. However, most cases report AFP levels that correspond to tumour activity,13 27 and relapses do occur.

Table 1

A summary table of 61 hepatocellular carcinoma cases of spontaneous regression

First author Publication year Age/sex Aetiology Recurrence pre-SR (y/n) Treatment prior to recurrence (duration) Presence of distant mets pre-SR (y/n) Bx primary pre-SR Bx metastasis pre-SR Pathology post-SR Bx confirming HCC at any point Treatment pre-SR Regression radiographically SR duration SR end point
Kimura29 2021 84F HCV No N/A No No N/A Liver: HCC with necrosis Yes No treatment Not specified N/A Curative liver resection
Kakuta30 2021 71M HBV Yes TACE+RFA Yes Not specified No Lung: HCC Yes Bladder cancer surgery and chemotherapy; HCC treated with TACE +RFA, resulting in SR of metastases PR 35 months Recurrence
Arjunan31




2021 65F ETOH No N/A No No N/A Not specified No No treatment Not specified 2 months Recurrence
53M HCV Yes TACE Yes Not specified Yes Not specified Yes No treatment Not specified 5 years Study end
53M NASH No N/A No No N/A Not specified No No treatment Not specified 9 years Study end
48M HBV Yes Liver resection Yes Yes Not specified Not specified Yes TACE resulting in SR of metastases Not specified 13 years Study end
62M None Yes Liver resection +sorafenib (2 years) Yes Yes Yes Not specified Yes Stereotactic RT to brain resulting in SR primary Not specified 11 years Study end
73M Cryptogenic No N/A No Yes N/A Not specified Yes No treatment Not specified 6 years Recurrence
Sonbare32 2020 74M NASH No N/A No No N/A Not specified No No treatment PR 2 years Study end
Costa-Santos33 2020 68M HCV Yes Liver resection No Yes N/A Not specified Yes Megesterol +Annona muricata +sofosbuvir/ledipasvir CR 5 years Study end
Kawaguchi34 2019 67M HCV Yes TACE+RFA No No N/A Not specified No SGLT2i×10 weeks CR Not specified Study end
Chohan35 2019 79F HCV No N/A Yes No No Not specified No No treatment PR 71 months Recurrence
Taniai36 2018 74M HBV Yes Liver resections No Yes (remote, 10 years prior) N/A Liver: no HCC, no necrosis Yes No treatment Not specified N/A Curative liver resection
Koya37 2018 83M HCV Yes TACE+RFA Yes No No Not specified No No treatment CR 2 years Study end
Alam28 2017 61M HCV No N/A No No N/A Not specified No No treatment PR 10 weeks Recurrence
Clos38 2017 72M ETOH No N/A No Yes N/A Liver: no HCC, only necrosis Yes No treatment PR 14 months Study end
Pectasides9 2016 53M HCV/ETOH No N/A Yes No No Thrombus: HCC Yes No treatment PR Not specified Recurrence
Gunasekaran39 2015 49M HCV Yes TACE +sorafenib (12 months) Yes No yes Not specified Yes Guyabano extract powder PR 13 months Study end
Seo40 2015 74F HCV Yes TACE Yes No No Not specified No High dose vitamin C PR 1 year Elective TACE
Yang27 2015 56M HBV No N/A No No N/A Liver: HCC No No treatment CR 14 years Recurrence
Okano41 2015 73M HBV No N/A No No N/A Not specified No No treatment CR 5 years Study end
Wang42 2015 50M HBV No N/A No No N/A Liver: HCC with necrosis Yes No treatment Not specified N/A Curative liver resection
Bhardwaj43 2014 74F None No N/A No Yes N/A Not specified Yes No treatment PR 1 year Study end
Jo44 2014 64F HBV No N/A No Yes (bx negative) N/A Liver: HCC with necrosis Yes No treatment PR 28 months Recurrence
Chiesara45 2014 65M NASH No N/A No Yes N/A Not specified Yes Aloe arborescens PR 1 year Study end
Okano46 2013 77M ETOH No N/A No No N/A Not specified No No treatment PR N/A Elective TACE
Sasaki47 2013 79M ETOH No N/A No No N/A Liver: HCC with necrosis Yes No treatment PR N/A Curative liver resection
Tomishige48 2013 76F None No N/A No Yes (bx negative) N/A Liver: no HCC, only necrosis No No treatment Not specified N/A Curative liver resection
Nakayama49 2012 92F Cryptogenic No N/A No No N/A Not specified No No treatment CR Not specified Study end
Harimoto50 2012 73M HCV Yes Liver resection Yes Yes No Not specified Yes No treatment CR 13 months Study end
Komatsu51 2012 65M HCV Yes No treatment. Multiple SR then recurrences. No No N/A Not specified No No treatment PR Not specified Elective TACE
Storey52 2011 52M ETOH No N/A Yes Yes Yes (bx inconclusive) Liver: no HCC, only necrosis Yes No treatment PR 15 months Curative liver resection
Okano53 2011 68M HCV Yes Liver resection No Yes N/A Not specified Yes No treatment CR 1 year Recurrence
Alqutub54 2011 65M None No N/A No No N/A Not specified No No treatment CR 2 years Study end
Oquinena7

2009 54M HBV No N/A No Yes (bx negative) N/A Not specified No No treatment CR 1 year Recurrence
61M ETOH Yes TACE No No No Not specified No No treatment CR 2 months Recurrence
60M Haemochromatosis No N/A No No N/A Not specified No Chemotherapy (six cycles) stopped months before SR CR 3 years Study end
Hsu55 2009 66M HCV No N/A No No N/A Liver: HCC with necrosis Yes Silymarin PR Not specified Curative liver resection
Meza-Junco56 2007 56F HCV No N/A No No N/A Liver: HCC with necrosis Yes No treatment PR N/A Curative liver resection
Kondo57


2006 70M HCV No N/A No No N/A Liver: HCC Yes No treatment PR 5 months Recurrence
75M HCV Yes TACE Yes No No Not specified No Chemotherapy (unclear duration) stopped months before SR PR 8 months Recurrence
67M HCV Yes RFA No Yes N/A Not specified Yes No treatment PR 1 month Died from another cause
67M HCV Yes PEI +RFA + liver resection Yes Yes No Not specified Yes ‘CAMs’ CR 5 years Study end
Kojima58 2006 79M HCV Yes PMCT+TACE Yes No No Not specified No P.linteus extract CR 10 months Study end
Ohta59 2005 74M None No N/A No No N/A Liver: no HCC, only necrosis No No treatment CR 1 year Curative liver resection
Ohtani60 2005 69M HCV No N/A No No N/A Liver: no HCC, only necrosis No No treatment PR 11 months Curative liver resection
Nam61 2005 65M HCV No N/A Yes Yes (bx negative) Yes Not specified Yes Mushroom Phellinus linteus +RT to skull resulting in SR of primary PR Not specified Study end
Jeon62 2005 72M None No N/A Yes No Yes Not specified Yes No treatment PR 6 months Study end
Nakai63 2001 76M HCV Yes liver resection +TACE No Yes N/A Not specified Yes No treatment CR 2 years Study end
Ikeda3 2001 75M HCV Yes TACE Yes No No Not specified No Chemotherapy (unclear duration) stopped months before SR PR 8 months Recurrence
Takeda64 2001 68M HCV Yes TACE+PEI No Yes N/A Not specified Yes Tahibo tea +Agaricus blazei murill CR 5 months Recurrence
Jang65 2000 54F HBV No N/A No Yes N/A Not specified Yes No treatment CR 4 years Study end
Castera66 2000 78M Haemochromatosis No N/A No Yes N/A Not specified Yes Tamoxifen CR 23 months Study end
Uenishi25 2000 65M HCV No N/A No No N/A Liver: HCC with necrosis Yes No treatment PR N/A Curative liver resection
Toyoda67 2002 82M HCV No N/A Yes No No Liver: HCC Yes No treatment PR Not specified Death from HCC
Misawa68 1999 62M HBV No N/A No No N/A Liver: no HCC No No treatment CR Not specified Study end
Ohba22 1998 76M HCV Yes Liver resection +TACE+PEI Yes Yes No Not specified Yes RT to bone mets resulting in SR of primary PR Not specified Study end
Sanz69 1998 66M HCV Yes Liver resection Yes Yes Yes Not specified Yes No treatment CR 5 years Study end
Kaczynski70 1998 73M None No N/A No Yes N/A Liver: no HCC Yes No treatment CR 15 years Death cause unknown
Van Halteren71 1997 72F None No N/A No Yes N/A Not specified Yes No treatment PR 28 months Study end
Iwasaki18 1997 72F HCV Yes TACE+PEI No Yes N/A Not specified Yes No treatment PR Not specified Study end
  • Bx, biopsy; CAMs, complementary and alternative medicines; CR, complete response; ETOH, alcoholic cirrhosis; F, female; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; M, male; Mets, metastases; N, no; N/A, not applicable; PEI, percutaneous ethanol injection; PMCT, percutaneous microwave coagulation therapy; PR, partial response; RFA, radiofrequency ablation; RT, radiation therapy; SGLT2i, sodium-glucose cotransporter-2 inhibitors; SR, spontaneous regression; TACE, transarterial chemoembolisation; Y, yes.

In summary, we present the case of a middle-aged man with presumed metastatic HCC who experienced enduring, partial regression following 3 weeks of lenvatinib, which was discontinued due to side effects. This could represent an unusually successful response to 3 weeks of targeted therapy, SR or an alternative diagnosis. To our knowledge, our patient is unique in experiencing 2 years of partial regression following radiation and short systemic treatment. Despite many attempts, a pathological confirmation has been unsuccessful. An unclear cancer diagnosis with unexpected regression has profoundly impacted our patient. While traditionally, the diagnosis of HCC can be made radiographically based on the The American Association for the Study of Liver Diseases (AASLD) criteria, the potential harm of not having a clear diagnosis could be detrimental to a patient’s well-being. We therefore advocate for tissue confirmation in the diagnosis of HCC in select cases, where there is diagnostic doubt or when the disease pattern does not clearly follow the recognised natural history. Therefore, if regression occurs—whether spontaneous or in response to treatment—it can be better understood and subsequent therapies recommended if needed.

Patient’s perspective

I have been confused. When I was first seen, I was told I had less than 6 months to live. I had a living wake and got my affairs in order. That was two years ago. Then everything got better. But I have never known do I have cancer or not. This unknown has made me emotionally wrecked to the point I needed antidepressants. Not knowing is the worst thing of this whole journey, even with the news that whatever I had is better.

Learning points

  • Hepatocellular carcinoma is a leading cause of cancer-related mortality and is associated with a poor prognosis. Developing targeted molecular and biological therapies have radically changed the treatment landscape.

  • Spontaneous tumour regression is a rare phenomenon in the field of oncology. An increasing number of reports describe this phenomenon in hepatocellular carcinoma. Often, the cancer diagnosis and spontaneous tumour regression (SR) are not confirmed histologically, which may result in management uncertainty.

  • Recurrences following SR occur, especially when the radiographical response is incomplete, so subsequent treatment is recommended when possible.

  • We advocate for tissue confirmation in the diagnosis of hepatocellular carcinoma in select cases, where there is diagnostic doubt or when the disease pattern does not clearly follow the recognised natural history.

Ethics statements

Patient consent for publication

Footnotes

  • Contributors LC and RR made equal and significant contributions to project and manuscript development, data collection, data analysis, result dissemination and knowledge translation. WL provided consulting expertise to manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Use of this content is subject to our disclaimer