Lenvatinib and pembrolizumab plus FOLFOX-hepatic arterial infusion chemotherapy in locally advanced, potentially resectable hepatocellular carcinoma

Abstract

BACKGROUND

Systemic therapies have improved management of hepatocellular carcinoma (HCC); however, further improvements in overall survival for first-line treatment of primary unresectable HCC remain necessary.

AIM

To evaluate the safety and efficacy of hepatic arterial infusion chemotherapy combined with lenvatinib and pembrolizumab for treating primary unresectable HCC.

METHODS

This retrospective study analyzed data from 38 patients who received this triplet therapy at Sun Yat-sen Memorial Hospital between December 2019 and March 2022.

RESULTS

The overall response rate was 50.0% (Response Evaluation Criteria in Solid Tumors criteria) and 52.6% (modified Response Evaluation Criteria in Solid Tumors criteria), with a disease control rate of 97.4%. The successful conversion rate, defined as meeting all prespecified criteria for surgical resectability was 50.0%. Consequently, the surgical conversion rate, reflecting the proportion of patients who underwent subsequent radical resection, was 44.7%. The pathological complete response rate was 7.9%. Grade III/IV treatment-related adverse events occurred in 10.53% of patients, with no perioperative deaths or major complications. The successful conversion group demonstrated significantly higher 1-year, 2-year, and 3-year survival rates compared to the unsuccessful conversion group.

CONCLUSION

Combination of hepatic arterial infusion chemotherapy, lenvatinib, and pembrolizumab is a promising and well-tolerated treatment strategy for primary unresectable HCC.

Key Words: Hepatocellular carcinoma; Conversion therapy; Transhepatic arterial perfusion chemotherapy; Lenvatinib; Pembrolizumab

Core Tip: This retrospective study of 38 patients with initially unresectable hepatocellular carcinoma treated with hepatic arterial infusion chemotherapy, lenvatinib and pembrolizumab showed promising results. The overall response rate was 50.0%-52.6% with a disease control rate of 97.4%. The conversion to surgery rate was 44.7%, with a pathological complete response rate of 7.9%. Treatment was well-tolerated, with only 10.5% experiencing grade III/IV adverse events. Patients achieving successful conversion had significantly better long-term survival.

INTRODUCTION

Primary liver cancer is the third leading cause of cancer deaths globally and the fourth most common malignant neoplasm. In China, it is the second major cause of cancer deaths, and 75%-85% of cases are hepatocellular carcinoma (HCC)[1,2]. Radical surgery is the most effective method for long-term survival in HCC patients. However, due to the slow development and late detection of HCC, < 30% of cases are eligible for surgery at an early stage[3].

Stage-reduction therapy aims to shrink advanced HCC to increase the rate of surgical R0 resection. However, due to the systemic nature of middle- and advanced-stage HCC and the limitations of transcatheter arterial chemoembolization (TACE), increasing surgical resection rates alone does not significantly improve survival rates. This led to the development of conversion therapy.

Conversion therapy transforms unresectable liver cancer into a form that can be surgically removed, offering a chance for long-term survival. Unresectable liver cancer has two levels: (1) Surgical unresectable, where patients cannot tolerate surgery due to overall health, liver function, or insufficient remaining liver volume (future liver remnant); and (2) Oncologically unresectable, where the tumor can be removed, but surgery offers no better outcome than nonsurgical treatments. Research shows that conversion therapy followed by stage II surgical resection can increase the 5-year survival rate for primary unresectable HCC to 24.9%-57.0%. This is comparable to the 30%-60% 5-year survival rate for primary resectable HCC[4]. Converting primary unresectable HCC into resectable is a pressing challenge and a key focus in HCC treatment research.

The main therapeutic options for HCC transformation are surgery, vascular intervention, radiation, and targeted immunotherapy. Due to the low success rate of single treatments, combining different methods is gaining attention. Triple therapy combining artery-directed therapy[5-7], antiangiogenic drugs, and immune checkpoint inhibitors (ICIs) has shown significant clinical effectiveness[8-11]. This study retrospectively evaluated the effectiveness of combining hepatic arterial infusion chemotherapy (HAIC), lenvatinib, and pembrolizumab for treating primary unresectable HCC at the author’s hospital. The aim of this study was to evaluate the real-world efficacy and safety of HAIC combined with lenvatinib and pembrolizumab as conversion therapy for primary unresectable HCC, with a focus on tumor response, conversion-to-resection rate, perioperative outcomes, and survival benefit.

MATERIALS AND METHODS

Medical information

A retrospective analysis was conducted on patients with primary unresectable HCC treated with HAIC, lenvatinib, and pembrolizumab in the Department of Hepatobiliary Surgery at Sun Yat-sen Memorial Hospital, from December 2019 to March 2022 (Figure 1). The study followed the diagnostic criteria from the Guidelines for the Diagnosis and Treatment of Primary Liver Cancer (2022 Edition)[2]. Primary unresectable HCC was determined based on the following criteria: (1) The tumor was large and affected major intrahepatic ducts; (2) The tumor was in an advanced stage and accompanied by cancerous clots in major blood vessels, such as the main trunk or first-grade branches of the portal vein, and the inferior vena cava; and (3) The remaining liver volume after tumor removal was inadequate, posing an extremely high risk of postoperative hepatic failure.

Figure 1  Flow diagram of study.

Inclusion criteria: (1) Primary unresectable HCC and potentially R0 resectable as assessed by the investigator; (2) Age 18-75 years; (3) Child-Pugh ≤ 7 points; (4) Eastern Cooperative Oncology Group-Physical Status 0-1 points; and (5) At least one measurable lesion [Response Evaluation Criteria in Solid Tumors (RECIST) criteria].

Exclusion criteria: (1) Important organ insufficiency; (2) Severe bleeding and coagulation dysfunction; (3) Multiple liver tumor lesions distributed in the left and right half of the liver; (4) Cancerous thrombus in the main trunk or first level branch of the portal vein; (5) Cancerous thrombus in the inferior vena cava; (6) Extrahepatic metastasis; (7) Indocyanine green 15-minute retention rate ≥ 15%; (8) Contraindication to the use of lenvatinib and pembrolizumab, such as autoimmune diseases; and (9) Underwent other systemic therapies. The study involved 38 patients who received HAIC, lenvatinib, and pembrolizumab.

Conversion treatment

The interventional therapy was HAIC with the FOLFOX4 regimen, which included oxaliplatin (85 mg/m²) on day 1, calcium folinic acid (200 mg/m²) on days 1 and 2, and fluorouracil (400 mg/m²) on days 1 and 2, plus a 40-hour continuous infusion of fluorouracil (1000 mg/m²) via the hepatic artery on day 1. The lenvatinib and pembrolizumab treatment protocol included 200 mg pembrolizumab intravenously every 21 days, and lenvatinib taken orally at 8 mg/day for those < 60 kg, or 12 mg/day for those ≥ 60 kg. The triple regimen was administered every 3 weeks. Comprehensive assessments, including physical examinations and laboratory tests such as blood pressure measurement, evaluation of lower limb swelling, hand-foot syndrome, blood analysis, liver and kidney function tests, thyroid function tests, cardiac enzyme profiling, urinalysis, and tumor marker measurements, were conducted every 3 weeks. Every 6 weeks, chest computed tomography (CT) and enhanced abdominal scans were performed to assess tumor response and surgical eligibility. Lenvatinib was stopped upon achieving complete response (CR) or partial response (PR), with hepatectomy performed 2 weeks later if the patient was eligible for R0 resection. The treatment was halted for grade III/IV adverse events (AEs) or tumor progression. For grade 2 AEs, the dosage was reduced or paused until the event subsided to grade 1, following the drug guidelines.

Medical procedures

Patients who have undergone successful conversion choose different radical resection techniques based on clinical factors such as number, size, and location of HCC lesions, remaining liver volume, and functional status of vital organs.

Clinical efficacy is evaluated based on the success rate of conversion, surgical conversion rate, pathological CR (pCR), objective response rate (ORR), 1-year survival rate, and 1-year liver cancer recurrence rate. For conversion to be considered successful, it must satisfy the following five conditions simultaneously: (1) CR and PR can be assessed using RECIST or modified RECIST (mRECIST) criteria; (2) R0 resection refers to achieving complete removal of the tumor with wide margins and regression of the cholangiocarcinoma embolus; (3) After R0 resection, it is important to ensure sufficient hepatic function, which means that the remaining liver volume should be > 40% of the standard hepatic volume and the indocyanine green 15-minute retention rate should be ≤ 20%; (4) Hepatic function should be assessed using the Child-Pugh score, which should be ≤ 7 points, and the Eastern Cooperative Oncology Group Performance Status score, which should be 0-1; and (5) Surgery should only be considered if there are no extrahepatic lesions and no other contraindications, such as impairment of vital organ function due to systemic therapy[12]. The surgical conversion rate is the ratio of patients who underwent successful conversion therapy and radical resection of HCC to the total number of patients receiving conversion therapy. Measurable lesions are > 1 cm on CT or magnetic resonance imaging, and target lesions are the largest and most proliferative liver cancer lesions. Patients are categorized by the total length of the longest diameters of up to five target lesions. In the RECIST criteria, CR is defined as the disappearance of all target lesions for ≥ 4 weeks. PR is defined as a reduction in the sum of the longest diameter of target lesions by ≥ 30% for ≥ 4 weeks. Progressive disease (PD) is defined as an increase in the sum of the longest diameter of target lesions by ≥ 20% or the appearance of new lesions. Stable disease (SD) is defined as a growth of the longest diameter of the target lesion by > 20% or a reduction of < 30%[13]. The pCR describes the absence of cancer cells in the specimen after surgical removal, as confirmed by pathological investigation. The mRECIST scoring system defines CR as the absence of blood flow in any blood vessels within the tumor, and PR as the absence of blood flow in the longest dimensions of the tumor during the arterial phase of contrast imaging. In this study, the term PR was used to describe a decrease of ≥ 30% in the total length of the largest diameters of the target lesions, which lasted for > 4 weeks. PD was defined as an increase of ≥ 20% in the total length of the largest diameters of the active target lesions. SD referred to any case that did not meet the criteria for PR or PD[14]. The ORR was the percentage of cases with complete or partial remission. The 1-year survival rate was the proportion of patients surviving > 1 year after conversion therapy and hepatectomy. The 1-year liver cancer recurrence rate was the percentage of patients who experienced recurrence within 1-year post-surgery. Treatment-related AEs (TRAEs) included indicators such as liver and renal function, blood cell counts, hypertension, proteinuria, hand–foot syndrome, gastrointestinal reactions, and immune-related conditions such as pneumonia and hepatitis. These indicators were assessed in accordance with the National Cancer Institute Common Terminology Criteria for AEs 5.0.

Definition of successful conversion and surgical decision-making

The primary outcome of successful conversion was stringently defined as meeting all the prespecified criteria listed in Table 1. These criteria were assessed and confirmed by a multidisciplinary team (MDT) comprising hepatobiliary surgeons, interventional radiologists, and medical oncologists. The decision to proceed with surgery was based on the concurrent fulfillment of all criteria following conversion therapy. The surgical conversion rate was defined as the proportion of patients who, after meeting all criteria for successful conversion, subsequently underwent radical hepatectomy.

Table 1 Criteria for defining successful conversion and surgical decision-making.

Category	Criterion	Definition/threshold	Assessment method and timing
Overall requirement	MDT consensus	Agreement by hepatobiliary surgery, interventional radiology, and oncology	MDT review after every 2 cycles of therapy
Radiological response	Objective tumor response	Achieve complete or partial response according to RECIST 11 and/or modified RECIST	Contrast-enhanced CT/magnetic resonance imaging every 6 weeks; response must be confirmed on two consecutive scans ≥ 4 weeks apart
	Technical resectability	Tumor downstaging to allow for a potential R0 resection. Regression of major vascular involvement (e.g., portal vein tumor thrombus)	Volumetric and anatomical assessment via imaging
Liver function and volume	FLR volume	FLR > 40% of standardized total liver volume	CT volumetry prior to surgical decision
	Liver function reserve	Child-Pugh score ≤ 7	Laboratory and clinical evaluation every 3 weeks
		ICG 15-minute retention rate ≤ 20%	ICG clearance test prior to surgical decision
Patient general status	Performance status	Eastern Cooperative Oncology Group Performance Status score: 0-1	Clinical assessment every 3 weeks
Systemic disease control	Extrahepatic metastasis	Absence of new or progressive extrahepatic metastases	Chest CT and baseline extrahepatic imaging every 6 weeks
Surgical safety	Major organ function	No severe, uncontrolled comorbidities precluding major surgery (cardiopulmonary, renal insufficiency, etc.)	Comprehensive preoperative anesthesiology evaluation

CT: Computed tomography; FLR: Future liver remnant; ICG: Indocyanine green; MDT: Multidisciplinary team; RECIST: Response Evaluation Criteria in Solid Tumors.

Follow-up and survival

We conducted follow-up examination via WeChat, telephone, or outpatient visits, obtaining complete information for all cases. Cases that expired due to unrelated causes or were unreachable during the follow-up period were considered as instances of tail amputation. Survival time was defined from the day of surgery to death caused by HCC. The number of surviving patients was recorded at the last follow-up, with a total follow-up period of 3 years.

Statistical analysis

The data processing and statistical analysis were conducted using SPSS version 26.0 statistical software. The numerical data were presented as rates, n (%), and the χ² test or exact probability approach was used. Analysis of ordered categorical data was conducted using a rank sum test. The measurement data was presented as mean ± SD and evaluated using independent samples t-test or analysis of variance. We used survival analysis using the Kaplan-Meier method, and the differences were assessed using the log-rank test. P < 0.05 indicated a significant difference.

RESULTS

Characteristics of patients at the beginning of the study

There were 38 patients who received combined treatment with HAIC, lenvatinib, and pembrolizumab, from December 2019 to March 2022. There were 36 males and two females. The age at the midpoint of the distribution was 53 years, with a range of 49-73 years. Thirty-four patients had a previous occurrence of hepatitis B and were not receiving consistent antiviral treatment (8 were receiving treatment, while 30 were not). Additionally, 47.4% of the patients (18 of 38) had cirrhosis. The distribution of HCC stages in the China Liver Cancers (CNLC) staging dataset was as follows: (1) Ia, one case; (2) Ib, five cases; (3) IIa, three cases; (4) IIb, one case; and (5) IIIa, twenty-eight cases. The Barcelona Clinic Liver Cancer (BCLC) classification system categorized HCC into three stages: (1) Early stage (A) with six cases; (2) Intermediate stage (B) with four cases; and (3) Advanced stage (C) with 28 cases. The range of HCC foci varied from one to five, with an average of 1.76. The average total of the longest diameters of the target lesions was 100.89 ± 38.13 mm (range 88.19-113.59 mm). There were 21 cases where the alpha-fetoprotein (AFP) level was ≥ 400 mg/L. Additional fundamental qualities are shown in Table 2.

Table 2 Baseline characteristics of 38 patients who had primary unresectable hepatocellular carcinoma.

Clinical characteristic	Data
Sex
Male	36
Female	2
Age (years)
≤ 65	32
> 65	6
Hepatitis B
Yes	34
No	4
Cirrhosis
Yes	18
No	20
Antiviral treatment
Yes	8
No	30
China Liver Cancers stage
Ia	1
Ib	5
IIa	3
IIb	1
IIIa	28
Barcelona Clinic Liver Cancer stage
Early (A)	6
Intermediate (B)	4
Advanced (C)	28
Number of cancer lesions
1	16
≥ 2	22
Sum of longest diameter of target lesions (cm)
≥ 10	23
< 10	15
Alpha-fetoprotein (μg/L)
≥ 400	21
< 400	17

According to RECIST criteria, there were three cases of CR, 16 of PR, 18 of SD, and one of PD. The ORR was 50%, disease control rate (DCR) was 97.4%, conversion success rate was 50%, surgical conversion rate was 44.7%, and pCR rate was 7.9%. Based on mRECIST criteria, there were five cases of CR, 15 of PR, 17 of SD, and one of PD. The ORR was 52.6%, DCR was 97.4%, conversion success rate was 52.6%, surgical conversion rate was 44.7%, and pCR rate was 7.9%. Although the ORR and conversion success rate were slightly higher with mRECIST than with RECIST, the difference was not statistically significant (P = 0.8185; Figure 2). Postoperative pathology confirmed a pCR in all three patients assessed as achieving a CR by RECIST criteria. In contrast, among the five patients assessed as CR by mRECIST criteria, pathological examination revealed residual viable tumor cells in two (40%), which were evaluated as PR by RECIST and failed to reach pCR, indicating a notable discrepancy between radiological and pCR in this subset. This highlights the potential limitation of mRECIST in predicting true pathological necrosis in the context of this combination therapy.

Figure 2 Comparative analysis of objective response rate and conversion success rates using Response Evaluation Criteria in Solid Tumors and modified Response Evaluation Criteria in Solid Tumors assessment criteria. mRECIST: Modified Response Evaluation Criteria in Solid Tumors; ORR: Objective response rate; RECIST: Response Evaluation Criteria in Solid Tumors.

Seventeen patients successfully underwent liver resection for HCC, including 13 Laparoscopic and four open operations, all under general anesthesia. Thirteen had anatomical hepatectomy, and four had irregular hepatectomy. The procedures included seven right hemihepatectomies, three left hemihepatectomies, four right posterior lobectomies, one right anterior lobectomy, one middle lobectomy, and one hepatectomy of segments S4 and S6. The median number of preoperative conversions was four (range 2-9). Evaluation of the effectiveness of conversion therapy (Table 3).

Table 3 Evaluation of effectiveness of conversion therapy (number of cases/percentage).

Assessment indicators	RECIST	Modified RECIST
CR	3/38 (7.9%)	5/38 (13.1%)
Partial response	16/38 (42.1%)	15/38 (39.5%)
Stable disease	18/38 (47.4%)	17/38 (44.7%)
Progressive disease	1/38 (2.6%)	1/38 (2.6%)
Objective response rate	19/38 (50.0%)	20/38 (52.6%)
Disease control rate	37/38 (97.4%)	37/38 (97.4%)
Pathological CR	3/38 (7.9%)	3/38 (7.9%)
Conversion success rate	19/38 (50.0%)	20/38 (52.6%)
Surgical conversion rate	17/38 (44.7%)	17/38 (44.7%)

CR: Complete response; RECIST: Response Evaluation Criteria in Solid Tumors.

Analysis of survival-related factors in conversion therapy

Univariate analysis of survival-related factors: Univariate analysis was conducted on factors including gender, age, hepatitis, liver cirrhosis, antiviral therapy, CNLC staging, BCLC staging, HCC lesion numbers, AFP levels, and conversion rate in 38 patients with initially unresectable HCC (Table 4). Factors with P < 0.2 affecting survival were included in multivariate Cox regression analysis. Results indicated that gender, hepatitis status, antiviral therapy, number of liver cancer lesions, and conversion success significantly influenced survival.

Table 4 Univariate analysis of survival factors in 38 patients with initially unresectable hepatocellular carcinoma, n (%).

Factors	Survival	Death	Hazard ratio (95%CI)	P value
Gender				0.114
Male	19 (52.8)	17 (47.2)	Reference
Female	2 (100)	0 (0)	3.297 (0.751-14.479)
Age (years)				0.725
≤ 65	18 (56.3)	14 (43.7)	Reference
> 65	3 (50.0)	3(50.0)	0.802 (0.236-2.734)
Hepatitis B				0.189
Negative	1 (25.0)	3 (75.0)	Reference
Positive	20 (58.8)	14 (41.2)	3.872 (0.514-29.169)
Liver cirrhosis				0.935
Negative	11 (55.0)	9 (45.0)	Reference
Positive	10 (55.6)	8 (44.4)	0.965 (0.41-2.274)
Antiviral therapy				0.093
Negative	14 (46.7)	16 (53.3)	Reference
Positive	7 (87.5)	1 (12.5)	2.179 (0.879-5.402)
China Liver Cancers stage				0.734
Ia	1 (100.0)	0	Reference
Ib	2 (40.0)	3 (60.0)	2.863 (0.362-22.623)	0.319
IIa	2 (66.7)	1 (33.3)	0.495 (0.113-2.172)	0.351
IIb	0	1 (100.0)	1.031 (0.236-4.498)	0.968
IIIa	16 (57.1)	12 (42.9)	0	0.983
Barcelona Clinic Liver Cancer stage				0.337
A	2 (33.3)	4 (66.7)	Reference
B	3 (75.0)	1 (25.0)	0.397 (0.09-1.746)	0.222
C	16 (57.1)	12 (42.9)	1.538 (0.442-5.353)	0.499
Hepatocellular carcinoma lesions				0.181
1	7 (43.7)	9 (56.3)	Reference
≥ 2	14 (63.6)	8 (36.4)	1.886 (0.745-4.776)
Alpha-fetoprotein (μg/L)				0.696
< 400	9 (52.9)	8 (47.1)	Reference
≥ 400	12 (57.1)	9 (42.9)	1.188 (0.5-2.823)
Conversion rate				0.011
Fail	9 (42.9)	12 (57.1)	Reference
Succes	12 (70.6)	5 (29.4)	0.272 (0.099-0.745)

Multivariate analysis of factors associated with survival: Univariate analysis identified five factors with P < 0.2: (1) Gender; (2) Hepatitis; (3) Antiviral therapy; (4) HCC lesion numbers; and (5) Conversion rate. These factors were incorporated into a multivariate analysis using a Cox proportional hazards model. Successful conversion was an independent factor affecting survival. Gender, hepatitis, antiviral therapy, and HCC lesion numbers had P > 0.05, indicating that they had no significant impact on survival (Table 5).

Table 5 Multivariate analysis of survival factors in 38 patients with initially unresectable hepatocellular carcinoma.

Factors	β	Wald	Hazard ratio	P value	95%CI
Gender				0.060	0.937-24.376
Male	-	-	Reference
Female	1.564	3.541	4.780
Hepatitis				0.107	0.691-44.754
Negative	-	-	Reference
Positive	1.715	2.599	5.559
Antiviral therapy				0.691	0.448-3.359
Negative	-	-	Reference
Positive	0.205	0.159	1.227
Lesion numbers				0.461	0.518-4.275
1	-	-	Reference
≥ 2	0.397	0.544	1.488
Conversion rate
Fail	-	-	Reference
Success	-1.609	8.263	0.200	0.004	0.067-0.599

TRAEs involving the transfer of research knowledge to clinical practice

Each patient experienced TRAEs: Four (10.53%) having grade III/IV events, including two cases of hypertension, one of proteinuria, and one of gastrointestinal bleeding. Among the 17 patients who underwent surgery, there were no perioperative deaths or major complications such as bile leaks, liver failure, ascites, or intra-abdominal infections. The study compared the efficacy of transformation in early (CNLC 1a-2a) and advanced (CNLC 2b-3a) HCC (Figure 3). The transformation success rate was higher in the early group (66.6%, 6/9 cases) than in the advanced group (44.8%, 13/29 cases), but the difference was not statistically significant (P = 0.447).

Figure 3  Comparative analysis of the success rate of transformation in early-stage and advanced-stage liver cancer.

Subsequent findings and examination of longevity

The follow-up deadline was March 2024, with an average follow-up of 12 months (range 1-36 months). The 1-year, 2-year, and 3-year survival rates in the successfully transformed group were 76.47%, 66.91%, and 66.91%, respectively, which were significantly higher than 38.09%, 21.76%, and 21.76% in the unsuccessful group (P = 0.0047; Figure 4). The combined use of HAIC, lenvatinib, and pembrolizumab significantly improved survival in patients with primary unresectable HCC. The 1-year recurrence rate after surgery was 17.6% (3 of 17 patients).

Figure 4  Kaplan-Meier overall survival curves for the successful and unsuccessful transformation groups.

Typical cases

Case information: In October 2021, a 29-year-old male patient weighing 46 kg was admitted with a hepatic mass detected during a physical examination, present for > 1 month. He had a 20-year history of chronic hepatitis B without regular treatment and no history of hypertension, diabetes, or heart disease. Laboratory results showed an AFP level of 121 mg/L and Child-Pugh class A liver function (6 points). Imaging revealed a large tumor in the right liver lobe, multiple intrahepatic lesions, liver capsule involvement, and cancerous clots in the right portal vein branch. The left adrenal gland showed slight hyperplasia, and the spleen was enlarged (consent for publication has been obtained from the patient for their case details to be published).

Translational treatment history: The patient presented with a large tumor in the right lobe of the liver measuring 182 mm × 139 mm × 132 mm, along with several smaller tumors. This tumor was diagnosed as a massive HCC with multiple intrahepatic metastases, classified as stage IIIa according to the CNLC staging system and BCLC-C according to the BCLC staging system (Figure 5). Following MDT debate, it was determined that conversion therapy would be conducted, which involved combining targeted immunity with hepatic artery perfusion chemotherapy. The precise plan consisted of HAIC-FOLFOX4, which involved administering oxaliplatin at 85 mg/m² on day 1, calcium folinate at 200 mg/m² on days 1 and 2, and fluorouracil at 400 mg/m² on day 1. Additionally, a continuous infusion of fluorouracil at 1 g/m² over 40 hours via the hepatic artery was used as part of the interventional treatment plan. The specific combination immunotherapy regimen consisted of pembrolizumab (Keytruda) administered intravenously at 200 mg every 21 days, and lenvatinib orally at 8 mg once daily. Based on the changes in the tumor, a subsequent surgical assessment was conducted.

Figure 5 Enhanced computed tomography findings before transformation (primary treatment). A and B: Postoperative computed tomography (CT) findings showed multiple nodules in the right liver lobe, measuring 46 mm × 44 mm × 53 mm in total. The intrahepatic mass and nodules were reduced in size compared to the previous scan, indicating ongoing tumor activity. The right portal vein and hepatic vein showed minor constriction due to external pressure. Imaging assessment indicated a partial response; C: Alpha-fetoprotein levels; D: Variation curve of the maximum diameter of the tumor; E: Gross specimen after surgery. A liver tissue sample (12.5 cm × 12 cm × 6.5 cm) was cut, revealing multiple nodules ranging from 0.6 cm to 6.5 cm in size. These nodules were greyish-yellow, solid, soft, and had distinct borders. They were located near the periosteum and did not reach the cautery margin; F: Postoperatively, the right liver lobe showed changes with a remaining cavity in the operated area, accompanied by minor fluid and gas around the liver and cavity. The area was being drained; G: One month after surgery, the right liver lobe was completely absent, with a partial wedge-shaped absence in hepatic S4. There was swelling near the S4 margin and a small amount of fluid around the liver; H: Three months after surgery, the right liver lobe and S2 showed postoperative changes, including a residual cavity at the surgical site. The gallbladder was absent, and fluid around the liver had been absorbed. A new abnormal growth in liver S2 suggested metastasis, warranting magnetic resonance imaging with liver-specific contrast (Prometrium) for further diagnosis. Additionally, multiple lung nodules had grown, indicating metastatic tumors; I: Six months after surgery, the right liver lobe and S4 showed postoperative changes with a residual cavity, and the gallbladder remained absent. A mildly enhancing nodule in liver S2 had grown slightly, suggesting a high likelihood of metastatic tumor. The spleen remained enlarged, and cranial CT showed no clear abnormalities; J: One year after surgery, the right liver lobe and S4 showed postoperative changes with a residual cavity, and the gallbladder was still absent. A mildly enhanced nodule in liver S2 had grown and showed less enhancement, suggesting a metastatic tumor. Other nodules did not show clear findings. The spleen remained enlarged, and cranial CT revealed no evident abnormalities; K: Two years after surgery, the right liver lobe and S4 showed postoperative changes with a residual cavity, and the gallbladder remained absent. The liver had several metastases; most of which had invaded the middle and right hepatic veins. Both lungs had multiple cancerous nodules, with most increasing in size. Lymph node metastases in the mediastinum and right hilum had also enlarged, with invasion of the single vein and compression of the right upper pulmonary artery. The spleen was enlarged, the main portal vein widened, the splenic vein thickened, and varicose veins developed in the lower esophagus and stomach fundus.

The patient underwent six sessions of HAIC combined with targeted immunotherapy on October 18, 2021, November 8, 2021, November 29, 2021, January 3, 2022, January 24, 2022, and February 28, 2022, along with antiviral hepatoprotective medication and regular monitoring. The patient had no significant discomfort or adverse symptoms. In April 2022, after the sixth treatment, abdominal CT showed multiple nodules in the right liver lobe, with a total size of 46 mm × 44 mm × 53 mm (Figure 5A and B). The nodules had decreased in size but remained active, with modest constriction of the right portal and hepatic veins. Imaging indicated a PR. AFP and the variation curve of the maximum diameter of the tumor gradually decreased after six cycles of HAIC combined with targeted immunotherapy (Figure 5C and D). The AFP level was 10075 μg/L, and liver function was Child-Pugh class A (5 points). An MDT discussion evaluated the potential for surgical resection post-conversion.

Surgical treatment history: In April 2022, after thorough preparation, we performed fluorescence laparoscopic right hemihepatectomy, left liver tumor excision, cholecystectomy, abdominal adhesion release, and hepatic hilar lymph node dissection (Figure 5E). Postoperative pathology revealed multiple moderately to poorly differentiated HCC with necrosis, hyaline cells, and invasion into the liver without breaching the hepatic peritoneum. Cancerous emboli were found in vessels, but the margins were clear. Liver fibrosis was staged as S2, with positive histochemistry for CD34, glypican-3, and AFP. Another liver mass showed extensive necrosis with residual moderately differentiated carcinoma invading the hepatic peritoneum but not breaching it, with no cancerous emboli or carcinoma in the margins, consistent with post-treatment changes (Figure 5F).

Monthly postoperative follow-up: Immunotherapy was given on June 30, 2022, July 21, 2022, August 12, 2022, and September 7, 2022. Regular follow-up abdominal CT scans were performed at 1 month, 3 months, and 6 months, and 1 year and 2 years postoperatively (Figure 5G-K). A 6-month postoperative CT confirmed metastatic lesions, leading to interventional therapy on June 2, 2022, and December 12, 2022. The patient received camrelizumab + apatinib + 5-fluorouracil + Zoya + oxaliplatin on January 6, 2023, January 31, 2023, April 11, 2023, May 12, 2023, June 12, 2023, September 2, 2023, and October 9, 2023, with no apparent AEs. The patient is being monitored and remains alive.

DISCUSSION

In China, the 5-year survival rate for HCC is about 12%. For surgically resectable cases, the rate is 30%-60%, while for nonresectable cases, it is < 20%[3,15]. Due to the insidious onset and rapid progression of HCC, 70%-80% of patients are diagnosed at an unresectable intermediate or advanced stage. Therefore, comprehensive evaluation of conversion therapy is crucial.

The formation of HCC is a complex process, involving nearly all carcinogenic pathways to some extent[16]. HCC involves multiple molecular pathways, including RAF/MEK/ERK, PI3K/AKT/mTOR, WNT/β-catenin, insulin-like growth factor, HGF/c-MET, and vascular growth factor signaling[17]. Abnormal vascular growth factor signaling, particularly elevated levels of vascular endothelial growth factor and fibroblast growth factor, significantly impacts HCC development. These elevated factors are linked to venous invasion and disease stage in HCC patients[18-20]. Lenvatinib is a kinase inhibitor targeting vascular endothelial growth factor receptors 1-3, fibroblast growth factor receptors 1-4, platelet-derived growth factor receptor alpha, RET, and KIT genes[21-24].

The development of targeted drugs like lenvatinib, regorafenib, and apatinib, along with immunotherapies like atezolizumab, pembrolizumab, and sintilimab, has advanced systemic therapy for HCC. This progress brings hope for conversion therapy, with combination immunotherapy showing synergistic effects, supported by promising randomized controlled trial results. The ORR for first-line treatment of advanced HCC was increased to 46% in several studies, including (1) The KEYNOTE-524 study of lenvatinib combined with pembrolizumab[5]; (2) Study 117 study of lenvatinib combined with nivolumab[25]; (3) The IMbrave150 study of atezolizumab plus bevacizumab[26]; and (4) The Leap002 study of pembrolizumab combined with lenvatinib. The ORR was 46.0%, 76.7%, 33.2%, and 28.1% respectively. Targeted combination immunotherapy has been successfully applied in the translational treatment of primary unresectable HCC, resulting in significant effectiveness. In a study published by the 2020 American Society of Clinical Oncology (Abstract, No. E16690), the team from Zhongshan Hospital of Fudan University utilized a combination of tyrosine kinase inhibitor (TKI) and programmed cell death protein (PD)-1 monoclonal antibody to treat 60 cases of initially unresectable advanced HCC. The surgical translational success rate was 18.3%. A group of researchers from Sun Yat-sen Memorial Hospital used a translational therapeutic approach using trenbolide monoclonal antibody in combination with bevacizumab to treat 11 patients with early-stage HCC that could not be surgically removed. Five patients had a PR, resulting in an ORR of 45%[12].

ICIs have significantly advanced cancer treatment[27]. Cytotoxic T-lymphocyte-associated antigen-4 and PD-1 are checkpoint proteins on cytotoxic T lymphocytes that interact with CD80/CD86 and PD ligand (PD-L) 1, respectively. These interactions reduce T-cell activity, helping cancer cells evade cytotoxic-T-cell attacks[28]. ICI inhibits the interaction between receptors and ligands, preventing cancer cells from evading death caused by cytotoxic T cells. This leads to an antitumor activity[29]. Hypoxia in the tumor environment aids cancer cells in evading immune detection by weakening immune effector cells. Activation of hypoxia-inducible factor 1α also increases PD-L1 expression in cancer, myeloid-derived suppressor, and dendritic cells[30]. Pembrolizumab, a humanized immunoglobulin G4-kappa monoclonal antibody, blocks the interaction between PD-1 and PD-L1. It has shown promising results in treating PD-L1-positive unresectable HCC[17]. Combining lenvatinib with pembrolizumab is effective because lenvatinib blocks new blood vessel formation in tumors, enhancing the immune response and boosting the effectiveness of PD-1 antibodies[31,32].

HAIC has advanced significantly and is now a key treatment for advanced HCC in Asia. HAIC improves tumor response rates and reduces systemic toxicity by targeting chemotherapeutic agents directly to tumor tissues[33]. Combining HAIC, TKI, and ICI in triple therapy enhances cancer inhibition and reduces tumor size[34-36].

Conversion therapy differs from simple tumor shrinkage and stage reduction in advanced HCC. Its goal is to reduce tumor size and stage and to enable complete tumor removal and long-term survival. Conversion therapy is characterized by its significant and timely effectiveness in reducing liver cancer to a resectable stage after three or four treatment cycles. Recent advances in local and systemic therapies have shown promising safety profiles and higher success rates in conversion, as demonstrated in multiple phase 3 trials[37].

Since the IMbrave150 trial, combining TKIs and ICIs has become a mainstream treatment for advanced HCC, achieving an ORR of 28.1%-38.6%[26]. With around 40% of IMbrave150 trial patients having prior TACE treatment, the trial has spurred the aggressive use of triple therapy – TACE/HAIC, ICIs, and TKIs – for unresectable HCC. Zheng et al[37] first reported the combined treatment of TACE, sorafenib, and ICIs in 22 patients with unresectable HCC. Triple therapy outperformed TACE plus sorafenib, with a higher DCR (81.82% vs 55.17%) and prolonged progression-free survival (16.26 months vs 7.30 months) and overall survival (23.3 months vs 13.8 months). This has led to growing interest in triple conversion therapy for unresectable HCC.

The study found that the HAIC, lenvatinib, and pembrolizumab regimen had high conversion efficiency, with an ORR of 50.0% (RECIST) and 52.6% (mRECIST), and a DCR of 97.4%. The conversion success rate was 50.0% and 52.6%, with a surgical conversion rate of 44.7%, consistent with previous studies (40.5%-46.8%)[38-41]. Univariate analysis was conducted on factors including gender, age, hepatitis, liver cirrhosis, antiviral therapy, CNLC staging, BCLC staging, HCC lesion numbers, AFP levels, and conversion rate in 38 patients with initially unresectable HCC (Table 4). Factors with P < 0.2 affecting survival were included in multivariate Cox regression analysis. Gender, hepatitis status, antiviral therapy, number of liver cancer lesions, and conversion success significantly influenced survival. Univariate analysis identified five factors with P < 0.2: (1) Gender; (2) Hepatitis; (3) Antiviral therapy; (4) HCC lesion numbers; and (5) Conversion rate. These five factors were incorporated into a multivariate analysis using Cox proportional hazards model. Successful conversion was an independent factor affecting survival. Gender, hepatitis, antiviral therapy, and HCC lesion numbers with P > 0.05, indicating they had no significant impact on survival.

Three cases achieved pCR. While ORR and conversion success rates were higher with mRECIST than RECIST, the difference was not significant. Postoperative pathology confirmed pCR in three RECIST cases, but residual cancer cells were found in two of five mRECIST CR cases. Lack of vascular enhancement on CT or magnetic resonance imaging after conversion therapy does not guarantee no remaining cancer cells. The 1-year, 2-year, and 3-year survival rates were significantly higher in the successful group, suggesting that patients with unresectable HCC should pursue triple conversion therapy to improve survival.

In 17 surgical cases, the median number of preoperative conversions was four, ranging from three to six. Experts agree that at ≥ 4 HAIC courses are needed for optimal conversion. A meta-analysis shows that most downstaging conversions occur within 6 months, during which regular evaluations for surgical eligibility are crucial.

The combined antitumor effects of HAIC, lenvatinib, and pembrolizumab involve several pathways. HAIC enhances systemic therapy by exposing tumor antigens, increasing tumor immunogenicity, and improving the immune microenvironment. Chemotherapeutic agents boost systemic therapy by raising leukocyte antigen expression, activating T cells, and restoring immune surveillance. Lenvatinib and pembrolizumab normalize blood vessels, disrupt the hypoxic environment of the tumor, and convert “cold” tumors into “hot” tumors, increasing immune activity.

Macrovascular invasion and extrahepatic metastases worsen HCC prognosis and are independent risk factors for conversion therapy failure. This study excluded HCC patients with combined portal vein trunk or first-degree branch cancer thrombus, inferior vena cava cancer thrombus, and extrahepatic metastasis. Since cancer thrombus responds well to radiotherapy, combining radiotherapy with other therapies may be better for unresectable HCC with macrovascular invasion. Further research on conversion therapy for HCC with these conditions is needed.

The primary concern in combination therapy is safety, as multiple drugs can increase side effects. In our conversion protocol, all patients experienced TRAEs, mostly mild to moderate (grade I/II). Only 10.53% (4 cases) had severe TRAEs, including hypertension (2 cases), proteinuria (1 case), and gastrointestinal bleeding (1 case). The low incidence of severe TRAEs suggests our conversion strategy is safe and well tolerated.

The patient initially received six HAIC treatments combined with targeted immunotherapy, followed by surgery after the disease was controlled. After surgery, the same treatment continued for four sessions. However, a CT scan 6 months later revealed disease recurrence. The treatment was switched to a regimen including dual aromatase inhibitors, 5-fluorouracil, L-nitrosamine, and oxaliplatin. This case highlights that, while improving transformation rates is crucial, postoperative recurrence remains a significant challenge in HCC treatment. Further research is needed to find the best strategies to reduce recurrence, improve quality of life, and increase survival, including determining optimal postoperative treatment plans and duration.

Our research found that CNLC 1a cases had a higher conversion success rate (66.6%) compared to CNLC 2b cases (44.8%), although this difference was not significant. This suggests that early to mid-stage cases with insufficient liver tissue respond better to triple conversion therapy, warranting further investigation into the underlying causes and mechanisms.

This study had several limitations that should be acknowledged. First, due to its retrospective, single-center, and single-arm design, the findings may be subject to selection bias and unmeasured confounding, and the absence of a parallel control group limits the ability to directly attribute outcomes to the triple regimen. To partially address this, we performed multivariable Cox regression analyses to adjust for key baseline clinical factors, yet residual confounding may persist. Second, the sample size was small, which may have reduced the statistical power of subgroup comparisons (e.g., CNLC stage-based analyses) and limited the generalizability of the results. Third, although response was evaluated using RECIST and mRECIST, radiological assessment may not fully reflect pathological response, and discrepancies between imaging-based responses and surgical pathology could not be completely addressed. In addition, treatment exposure, timing of surgery, and postoperative management may have varied among patients in real-world practice, which could have influenced long-term outcomes, particularly recurrence. Finally, the follow-up duration and event numbers were limited for comprehensive evaluation of survival and recurrence patterns. Therefore, further large-scale, prospective studies are warranted to validate these findings and to optimize patient selection, treatment sequencing, perioperative decision-making, and recurrence prevention strategies.

Overall, HAIC combined with lenvatinib and pembrolizumab demonstrated promising efficacy and a favorable safety profile as conversion therapy for patients with primary unresectable HCC. This triple regimen achieved high radiological response and a substantial conversion-to-resection rate, suggesting that it may offer a clinically meaningful opportunity for curative-intent surgery in selected patients. Importantly, patients who achieved successful conversion experienced improved survival compared with those who did not, supporting the potential value of this strategy in real-world practice. Future studies should further refine patient selection and optimize treatment sequencing, surgical timing, and postoperative management to maximize long-term benefit and reduce recurrence.

CONCLUSION

The concurrent use of HAIC with lenvatinib and pembrolizumab was safe and effective. This combination therapy shows promise as a potential preferred treatment approach for primary unresectable HCC.