Abstract
Background/aims
The aim of this study was to prospectively evaluate stereotactic body radiotherapy (SBRT) with robotic radiosurgery in hepatocellular carcinoma patients with macrovascular invasion (HCC-PVT).
Materials and methods
Patients with inoperable HCC-PVT, good performance score (PS0-1) and preserved liver function [up to Child-Pugh (CP) B7] were accrued after ethical and scientific committee approval [Clinical trial registry-India (CTRI): 2022/01/050234] for treatment on robotic radiosurgery (M6) and planned with Multiplan (iDMS V2.0). Triple-phase contrast computed tomography (CT) scan was performed for contouring, and gross tumour volume (GTV) included contrast-enhancing mass within main portal vein and adjacent parenchymal disease. Dose prescription was as per risk stratification protocol (22–50 Gy in 5 fractions) while achieving the constraints of mean liver dose <15 Gy, 800 cc liver <8 Gy and the duodenum max of <24 Gy). Response assessment was done at 2 months’ follow-up for recanalization. Patient- and treatment-related factors were evaluated for influence in survival function.
Results
Between Jan 2017 and May 2022, 318 consecutive HCC with PVT patients were screened and 219 patients were accrued [male 92%, CP score: 5–7 90%, mean age: 63 years (38–85 yrs), Cancer of the Liver Italian Program <3: 84 (40%), 3–6117 (56%), infective aetiology 9.5%, performance status (PS): 0–37%; 1–56%]. Among 209 consecutive patients accrued for SBRT treatment (10 patients were excluded after accrual due to ascites and decompensation), 139 were evaluable for response assessment (>2 mo follow-up). At mean follow-up of 12.21 months (standard deviation: 10.66), 88 (63%) patients expired and 51 (36%) were alive. Eighty-two (59%) patients had recanalization of PVT (response), 57 (41%) patients did not recanalize and 28 (17%) had progressive/metastatic disease prior to response evaluation (<2 months). Mean overall survival (OS) in responders and non-responders were 18.4 [standard error (SE): 2.52] and 9.34 month (SE 0.81), respectively (P < 0.001). Mean survival in patients with PS0, PS1 and PS2 were 17, 11.7 and 9.7 months (P = 0.019), respectively. OS in partial recanalization, bland thrombus and complete recanalization was 12.4, 14.1 and 30.3 months, respectively (P-0.002). Adjuvant sorafenib, Barcelona Clinic Liver Classification stage, gender, age and RT dose did not influence response to treatment. Recanalization rate was higher in good PS patients (P-0.019). OS in patients with response to treatment, in those with no response to treatment, in those who are fit but not accrued and in those who are not suitable were 18.4, 9.34, 5.9 and 2.6 months, respectively (P-<0.001). Thirty-six of 139 patients (24%) had radiation-induced liver disease (RILD) [10 (7.2%) had classic RILD & 26 (19%) had non-classic RILD]. Derangement in CP score (CP score change) by more than 2 was seen in 30 (24%) within 2-month period after robotic radiosurgery. Eighteen (13%) had unplanned admissions, two patients required embolization due to fiducial-related bleeding and 20 (14%) had ascites, of which 9 (6%) patients required abdominocentesis.
Conclusion
PVT response or recanalization after SBRT is a statistically significant prognostic factor for survival function in HCC-PVT.
Keywords: hepatocellular carcinoma, portal vein thrombosis, robotic radiosurgery, recanalization
Almost 30% of hepatocellular carcinoma (HCC) patients present with vascular invasion, either portal vein thrombosis (PVT) or inferior vena cava (IVC) invasion.1, 2, 3 A proportion of these patients are diagnosed with screening or with general abdominal symptoms and usually have preserved liver function status (Child-Pugh status). Vascular invasion in HCC is considered a poor prognostic factor [Barcelona Clinic Liver Classification (BCLC) C] and is usually treated with palliative intent systemic therapy or immunotherapy.6 Conventional focal (local) therapies such as surgery, radiofrequency ablation (RFA), trans-arterial chemoembolization (TACE) or transplant is not feasible in this situation.4,5 Patients with vascular invasion respond poorly to systemic therapies (tyrosine kinase inhibitors; TKIs) and usually survive for 6–9 months.6,7 Fortunately, even these patients with vascular invasion usually do not have metastatic disease in the initial period and hence mandate exploration with novel focal therapy approaches.8 Stereotactic radiosurgery (SBRT) with robotic radiosurgery (CyberKnife, CK) is considered as a novel focal therapy approach with high-dose radiation therapy, with an intention to recanalize vascular invasion.
We have reported in our previous publication about the interim analysis of the prospective study on HCC with vascular invasion treated with SBRT.9 Patients with recanalization (responders) have better survival mostly because after recanalization, liver decompensation is deferred and has a positive impact on survival function. ‘Non-responders’, those with portal vein thrombosis not responding to SBRT, have poorer survival due to early liver decompensation. This study presents the final analysis of the study after accrual of planned patient numbers for analysis. The intention of this study is to analyse of the study hypothesis with adequately powered sample size and long-term follow-up.
Materials and Methods
The present study is registered with the Indian Council of Medical Research (ICMR) clinical trial registration [CTRI: 2022/01/050234] and is an institutional ethical (AIMS/IRB/14/2017) and scientific committee review board–approved prospective phase II single-arm observational study. Details of the study protocol are explained in our previous interim analysis publication.9 Patients with good performance status (preferably up to PS1), preserved liver function (preferably up to CP B7) and radiologically confirmed HCC with vascular invasion were prospectively accrued in the study. Only a few patients with good PS and small volume disease with more than CP B7 score were treated conservatively with albumin, diuretics, peritoneal fluid drainage and Vit K supplementation and were accrued in the study after improvement of the CP score. In the demographic profile, however, the pre-conservative treatment score was reported. The primary endpoint of the treatment was to assess response to treatment (recanalization rate or shrinkage of thrombus). The secondary endpoint was to evaluate survival function, toxicity parameters and factors influencing the outcomes.
Treatment Details
The treatment details, dose prescription, follow-up pattern, response assessment approach and prognostication have been published earlier along with the interim results.9,10 After accrual and obtaining consent for the study, patients were planned for treatment as per the standard protocol.9 Patients underwent fiducial placement, CT simulation, contouring, planning and treatment delivery as per protocol. Contouring was done by radiation oncologist (DD) with inputs from a radiologist (RK). Standard organs at risk (OARs) structures (liver, duodenum, small bowel and kidney) were contoured.11 Planning was done with Multiplan (iDMS 2.0) planning system.
Response Assessment and Follow-up Parameters
Contrast-enhanced triple-phase computed tomography (CT) scan, liver function test, international normalized ratio (INR) and alpha feto protein (AFP) were evaluated at every follow-up. Toxicity assessment was done using Common Toxicity Criteria (CTCAE V4.03). Signs and investigation parameters were observed for classical radiation-induced liver disease (RILD) [Alkaline phosphatase (ALK) >2 times, ascites, anicteric hepatomegaly] and non-classical RILD (liver decompensation). Liver function test, AFP and blood counts were evaluated at every follow-up. Response assessment was done at 8–10 weeks post robotic radiosurgery with triple-phase-contrast CT scan. Response assessment was done by an interventional radiologist (RK). Response assessment was done based on response assessment criteria in solid tumour (RECIST), mRECIST and Positron Emission Tomography (PET) uptake (when done). Subjective criteria, shrinkage in tumour size as per target lesion criteria of RECIST, were also considered for response evaluation. Response was characterized as complete response (CR), progressive disease (PD) and stable disease. Patients with no contrast enhancement but persistent thrombus are considered ‘bland thrombus’. Overall survival (OS) functions were calculated from the date of registration to time of death or date of the last follow-up.
Prognostic Factor Evaluation
Prognostication was done based on Chen classification of PVT, CP score, AFP scoring, Cancer of the Liver Italian Program (CLIP) score and BCLC classification. Change in the CP score after treatment and its correlation with toxicity parameters were also evaluated.
Statistical Analysis
Statistical analysis was done using IBM SPSS 20.0 (SPSS Inc, Chicago, USA). Sample size was calculated considering response in 30% of patients, alpha error of 0.05 and power of 80%, with dropout rate of 10%. For the prospective study, sample size of 114 was calculated with confidence interval (CI) of 95%.9 Survival rates were calculated using Kaplan-Meier methods and compared using the log-rank test. A P-value <0.05 was considered as statistically significant.
Results
Consort Chart
From January 2017 until May 2022, 318 liver tumour patients were screened in the study represented in the consort chart (Figure 1). Among 318 screened patients, 219 (69%) were accrued for radiosurgery treatment and 98 (30%) were not considered for the treatment [patients eligible but not willing to participate in the study = 44 (12%), visceral metastasis = 17 (5%), poor GC [Eastern Cooperative Oncology Group (ECOG) PS > 2] = 15 (5%), CP score C with poor PS = 15 (4.9%), referred for TACE/RFA = 7 (3%)]. Among 219 patients considered for radiosurgery and who underwent planning CT simulation, 3 patients were cancelled due to ascites, and CT-guided fiducial placement was done in 216 patients by an interventional radiologist (SG). Among 216 patients who underwent fiducial placement, 4 patients (2%) had decompensation post fiducial implantation. Among these four patients, two had bleed within the liver and one patient required urgent embolization. One patient (0.5%) had cardiac arrest within 3 days after fiducial placement, 1 (0.5%) had pulmonary thromboembolism and 1 (0.5%) patient deferred treatment; these patients were not considered for SBRT. Hence, SBRT was delivered in 209 patients. Of them, 171 had HCC with PVT and 38 (18%) had HCC without macrovascular invasion. Among the 171 treated HCC with PVT patients, 28 (13%) had progression/metastasis within 2 months and were not eligible for response assessment and 139 patients were eligible for analysis.
Figure 1.
Consort diagram.
Demography
The demography profile of accrued patients (n = 209) is explained in Table 1. Mean age of presentation is 63 years (range: 38–85 years; male: 92%) (Table 1). Common presenting symptoms were abdominal pain (33%), fatigue (12%), variceal bleed (7%), abdominal distension (5%), jaundice (2%) and melaena (2%). It was diagnosed as a follow-up of chronic liver disease in 22% and incidentally in 17%. Twenty-three percent had diabetes, 4% had hypertension, 26% had both diabetes and hypertension, while 45% had no known co-morbidities. Infective aetiology (hepatitis B or C) was seen in 11 (9%). ECOG PS of 0 (PS0) was in 78 patients (37%), PS1 in 117 (56%) and PS2 in 14 (7%) patients. The baseline CP score was 5–7 in 189 (90%) and 8–10 in 19 (9%): CP A in 159 (76%), CP B in 45 (21%) and CP C in 4 (2%). [Patients with CP C score were treated with conservative management, and those with an improvement in the CP score (CP B7 or less) were accrued in the study.] Yerdel portal vein thrombosis classification was VP 1–2 in 35 (17%), VP 3 in 51 (24%), VP 4 in 85 (40%) and no vascular invasion in 37 (17%). The AFP score was <10 in 31 (15%), 10–150 in 59 (28%), 150–500 in 26 (12%) and >500 in 63 (30%). The CLIP score was 1–2 in 84 (40%) and 3–6 in 117 patients (56%).
Table 1.
Demography (n = 209).
| Criteria | Characteristics | n = 209 (%) |
|---|---|---|
| Age | Mean ± SD: 63.08 ± 8.8 years Median: 64 years Range: 38–85 years |
|
| Gender | Male | 192 (92) |
| Female | 17 (8) | |
| Co-morbidities | Nil | 93 (44.5) |
| Diabetes | 48 (23) | |
| Hypertension | 9 (4.3) | |
| Diabetes + hypertension | 55 (26.3) | |
| Others | 4 (1.9) | |
| ECOG PS | 0 | 78 (37.3) |
| 1 | 117 (56) | |
| 2 | 14 (6.7) | |
| Child-Pugh score | 5–7 | 189 (90.4) |
| 8–10 | 19 (9.1) | |
| Child-Pugh class | A | 159 (76.1) |
| B | 45 (21.5) | |
| C | 4 (1.9) | |
| PVT type | Vp1-2 | 35 (16.7) |
| Vp3 | 51 (24.4) | |
| Vp4 | 85 (40.6) | |
| 0 | 37 (17.7) | |
| AFP scoring system | <10 | 31 (14.8) |
| 10–150 | 59 (28.2) | |
| 150–500 | 26 (12.4) | |
| >500 | 63 (30.1) | |
| CLIP score | <3 | 84 (40.2) |
| 3–6 | 117 (56) |
AFP, alpha feto protein; CLIP, Cancer of the Liver Italian Program; PVT, portal vein thrombosis; SD, standard deviation.
Dosimetry Parameters
Among the 139 patients analysed, mean planning target volume (PTV) volume was 74.7 cc [standard deviation (SD): 53.7; PTV median: 56.5 cc, range: 10–340 cc] and mean PTV dose was 33.8 Gy (SD: 7.4; mean: 35 Gy, range: 22-50 Gy). Mean prescription isodose was 87%. Mean total liver volume was 1750 cc (SD: 466.23; median 1677 cc, range: 607–3583 cc). Mean liver dose was 11.2 Gy (SD: 2.45; median: 11.3 Gy, range: 6-20 Gy). Mean liver volume spared of 8 Gy (<800 cGy) was 831 cc (SD: 257 cc; median: 765 cc, range: 347–1811 cc). Mean duodenal max dose was 17.1 Gy (SD: 1.13; median: 18 Gy, range: 1.13–30 Gy). Mean small bowel max dose was 22 Gy (SD: 12.5; median: 19 Gy, range: 3.5–48.64 Gy) (Table SUPPL 1).
Outcome Parameters
The mean follow-up period was 12.21 months (SD: 10.66; range: 2–54). The follow-up period was more than 6 months in 82 (59%), more than 12 months in 32 (23%) and more than 24 months in 10 (7%). At the last follow-up, 51 (37%) were alive and 88 (63%) expired due to disease progression (Table 2). Of the 139 patients evaluated for response, 82 (59%) had response (recanalization) and 57 (41%) did not have response (no re-canalization). Among the 82 responders, 42 (30%) had bland thrombus, 27 (19%) had partial re-canalization and 13 (9%) had complete re-canalization (Table 2). Cause of death was local progression of disease in 77 (87%) patients, metastatic disease in 8 (9%), liver decompensation in 2 (2.5%) and cancer/treatment unrelated in 3 (3%) (1 cardiac arrest and two COVID-related deaths). Six patients (4.13%) underwent subsequent Trans-arterial chemoembolisation (TACE) or Trans-arterial radioembolisation (TARE) (Table SUPPL 2). The survival function was calculated by Kaplan-Meir analysis, and log-rank test was used to test the significance between responders and non-responders (Figure 2). Estimated mean OS for the entire cohort is 14.7 months (standard error: 1.62; 95% CI: 11.5–17.9 mo) (Fig SUPPL 1). Univariate analysis was done for factors such as age (OS: <63 yrs: >63 yrs = 11.3:14.6; P: 0.24), gender (OS: male: female = 12.7:13.7; P: 0.56), infective aetiology (OS: yes: no = 17.6:12.4; p: 0.165), disease pattern (OS: multi-focal <50%: >50% involved = 14.32:11.3; P: 0.1), AFP score (OS: 10–150: 150:500: >500 = 14.4:11.6:11.2;0.554), RT dose (OS: <34 Gy: >34 Gy = 10.9:13.7; 0.413), adjuvant TKIs (OS: received: not received = 13:12; 0.278), baseline CP score (OS: PS5-7: PS 8–10 = 12.9:14.9; P: 0.546), PVT score (OS: VP1-2: VP3 = 14.8:11.4; 0.740) and CLIP score (OS: <3:3–6 = 13.7:11.6; 0.208), and no factors were seen to influence survival function other than ECOG PS (OS: PS 0:1:2 = 16.99:11.7:9.7; P-0.019). PVT response to treatment (OS: responder: non-responder = 18.4:9.34; P < 0.001) had a significant correlation with survival function. Pattern of recanalization was also significant (OS: complete recanalization: partial recanalization: bland thrombus = 30.3:12.4:14.1; P: 0.002) (Figure 3).
Table 2.
Outcome Parameters.
| Parameters | n (%) | ||
|---|---|---|---|
| Follow-up (months) | Mean (SD) | 12.21 (10.66) | |
| ≥6 months | 82 (59) | ||
| ≥12 months | 32 (23) | ||
| ≥24 months | 10 (7.2) | ||
| Last follow-up | Alive | 51 (36.7) | |
| Dead | 88 (63.3) | ||
| Complete treatment | 171 | ||
| Not evaluable | Treated and yet not eligible for re-assessment | 4 | |
| Progression/metastasis | 28 | ||
| Evaluable for response assessment | 139 | ||
| PVT response to RT | Response | Re-canalization | 82 (59) |
| No response | No re-canalization | 57 (41) | |
| Type of response | Bland thrombus | 42 (30.2) | |
| Partial re-canalization | 27 (19.4) | ||
| Complete re-canalization | 13 (9.4) | ||
| Cause of death | Local progression | 77 (87.6) | |
| Metastatic | 8 (9) | ||
| Unrelated | 3 (3.4) | ||
| Subsequent TACE/TARE | 6 (4.13) | ||
PVT, portal vein thrombosis; RT, radiation therapy; SD, standard deviation; TACE, trans-arterial chemoembolization.
Figure 2.
Overall survival in responders & non-responders treated with robotic radiosurgery.
Figure 3.
Overall survival with a pattern of recanalization after robotic radiosurgery.
Estimated mean OS for responders is 18.4 mo (SE: 2.52; 95% CI: 13.45–23.34) and non-responders is 9.34 mo (SE: 0.81; 95% CI: 7.74–10.94) (P-value<0.001) (Figure 2). Six-month survival is 74% vs 66% and 12-month survival is 48.5% vs 24.5% in responders and non-responders, respectively (Table 3). Performance status was also an independent prognostic factor for survival functions (Fig SUPPL 2).
Table 3.
Survival Outcome – Responder, Non-responders, Deferred and Non-eligible Patients.
| Responders (n = 82) | Non-responders (n = 57) | P-value | Eligible, deferred (n = 16) | Non-eligible (n = 23) | |
|---|---|---|---|---|---|
| Mean estimated OS (SE) | 18.4 mo (2.52) | 9.34 mo (0.81) | 0.001 | 5.9 (2.08) | 2.6 (0.45) |
| 95% CI | 13.45–23.34 | 7.74–10.94 | 1.88–10.04 | 1.72–3.49 | |
| 6-month survival | 73.9 % | 66.3 % | 32.4% | 8.7% | |
| 12 month survival | 48.5 % | 24.5 % | 10% | 0% | |
| Median estimated OS (SE) | 12 mo (1.35) | 8 mo (0.77) | 3 (0.93) | 3 (0.73) | |
| 95% CI | 9.34–14.65 | 6.48–9.51 | 1.17–4.82 | 1.55–4.44 |
CI, confidence interval; OS, overall survival; SE, standard error.
Toxicity
Among the 139 patients, 18 (13%) had unplanned admissions within 2 months, 21 (15%) had one fiducial migration immediately after fiducial placement (<24 h) and 20 (14%) had ascites, of which nine patients required ascitic tapping, six required albumin infusion and three patients were managed conservatively with oral medication (diuretics) (Table SUPPL 2). Nine patients (6.5%) had liver decompensation, 8 (6%) had pain and 3 (2%) had bleed. On the whole, classic RILD was seen in 10 (7.2%) and non-classic RILD was seen in 26 (18.7%). Derangement in the CP score (CP score change) by more than 2 was seen in 24 (20%) within a 2-month period after CK (prior to response evaluation).
Comparison of Outcomes Between Patients Who Received SBRT and Those Who Deferred Treatment
Among 318 patients screened, 219 were planned for SBRT. Of the remaining 98 patients who did not receive SBRT, Forty-four patients were eligible but did not give consent for the study and 54 patients were not eligible for the study (poor PS, metastasis, CP C) (Table 4). Among the 98 patients not treated with SBRT, outcome data were available for 39 patients (44.8%) [deferred treatment = 16 (17.3%) and non-eligible = 23 (25%)]. Estimated mean OS of patients who had response to treatment was 18.4 months (95% CI: 13.45–23.34), that of non-responders was 9.34 months (95% CI: 7.74–10.94), that of those who deferred treatment was 5.9 months (95% CI: 1.88–10.04) and that of those who were non-eligible was 2.6 months (95% CI: 1.7–3.5) (Figure 4). Hence, eligible patients treated with SBRT had significantly higher survival function than patients not treated (Table SUPPL 3). The treatment plan and response assessment are shown in Figure 5.
Table 4.
Comparison Between Treatment Modalities in HCC With PVT.
| Treatment | Author, year | n | OS | Toxicity |
|---|---|---|---|---|
| Surgery: Hepatectomy |
Kenneth S H Chok, 20148 | 88 | Median OS: 10.9 mo. |
Grade ≥3: 14 (19.7%) |
| En block resection | 9.4 mo | |||
| Thrombectomy | 8.58 mo. | |||
| TACE | JP Sliva, 20179 | 1933 | Median OS: 8 (95% CI: 5–15) mo. 1-year OS: 29% |
1% liver failure; 18% other complications |
| TARE | V Vilgrain, 201710 | 143 | Median OS: 8 (95% CI: 6.7–9.9) mo. | Grade ≥3: 41% |
| Sorafenib | J Bruix, 201711 | 162 | Median OS: 6.1 mo. | Grade ≥3: 33% |
| Lenvatinib | M Kudo, 201812 | 109 | Median OS: 11.8 mo. | Grade ≥3: 75% |
| Regorafeniba | J Bruix, 201725 | 110 | Median OS: 10·6 months (95% CI 9·1–12·1) | Grade ≥3: 67% |
| Cabozantiniba | Abou-alfa GK, 2018 | 129 | Median OS: 7.6 mo. | Grade ≥3: 58% |
| Nivolumab + ipilimumaba | Thomas Yau, 2020 | 50 |
bObjective response: 27–32% Median OS: Arm A: 22.8 mo (95% CI: 9.4-not reached) Arm B: 12.5 mo. Arm C: 12.7 mo. |
Grade ≥3: 31–53% |
| SBRT (CK) | Present study, 2024 | 139 | Median OS: Responders: 18.4 (95% CI: 13.5–23.3) mo. Non-responders: 9.34 (95% CI: 7.74–10.94) mo. |
Classic RILD: 7.2% Non-classic RILD: 18.7% |
n = no of patients with macrovascular invasion in each study.
CI, confidence interval; CK, CyberKnife; HCC, hepatocellular carcinoma; PVT, portal vein thrombosis; OS, overall survival.
Post sorafenib progression.
Results of entire cohort (not PVT alone).
Figure 4.
Overall survival comparison among treated, untreated and unfit patients treated with robotic radiosurgery.
Figure 5.
Treatment plan (A&B) with robotic radiosurgery in Multiplan (iDMS) planning system and response assessment (C&D) 2 months post robotic radiosurgery follow-up.
Discussion
HCC with vascular invasion is considered as advanced stage disease (BCLC C) and is commonly treated with palliative systemic therapy (TKIs or immunotherapy).12 However, mean OS with sorafenib is only 3.1 months and almost 10% of patients have severe toxicity (Grade 3/4) (diarrhoea and skin rash) which limits the usage of this medicine. More than 40% of patients either reduce the dose due to toxicity or stop medication, which further diminishes the efficacy of these medicines. Propensity match analysis also suggested that there is no survival difference in HCC with PVT patients treated with or without sorafenib.7 TACE is not routinely recommended in apprehension of liver decompensation due to occlusion of hepatic arteries during TACE which is the only blood supply to the liver.12 Major portal vein thrombosis is a contraindication for transplant and also anatomical resection. Even after response to SBRT, these patients are not considered for transplant, unlike segmental portal vein involvement, and the outcome is dismal. Hence, patients who have major portal vein involvement (VP 3/4) need to be addressed differently.
In the present study, patients with major portal vein thrombosis (VP3/4) were accrued in a prospective study and treated with SBRT for the portal vein. Patients with recanalization after SBRT have shown promise to improve outcomes. Rationale for improving survival function after recanalization of portal vein is by improving hepatic blood flow through portal circulation, which stabilizes liver function or delays decompensation of liver. Delay in liver decompensation itself could lead to improved survival. Hence, even though the entire tumour is not targeted in large or multi-focal HCC only targeting the portal vein and adjacent tumour tissue, after response (recanalization) there is improved blood flow and delay in liver decompensation that in turn improves survival. Recanalization may improve possibility of curative intent treatment approach with RFA, TACE or TARE. Recanalization reduces portal hypertension and related complications. After recanalization, intervention for oesophageal varices and other portal hypertension–related complications are reduced. Quality of life, hospital admission and liver function improve, which in turn also improves OS. In fact, SBRT may be an opportunity to provide treatment options in HCC with major portal vein thrombosis, where no other available treatment options are practically useful. Higher biologically equivalent dose (BED) with SBRT may help in improving the response rate. In HCC with PVT patients, SBRT followed by TACE has shown promising results.13
Recanalization rate after SBRT in literature varies between 30 and 60% of patients depending upon various treatment-related, patient-related and assessment-related factors.
In recent years, few studies on the effectiveness and safety of SBRT for treating HCC have been published (Table 4). Considering the results of these reports, use of SBRT to target PVT may also have the potential to improve the survival outcomes by controlling the progression of the disease. Portal vein invasion is a relative contraindication for surgery. In recent years, advances in surgical techniques have made it possible to have safe resection of both hepatic tumours and PVT, with a peri-operative mortality rate of 6% (Table 4). Patients with Chen VP1/2 PVT have perioperative mortality of 3% with 5-Yr OS of 10–59%. However, in VP 3 & 4, the outcome is dismal (5-Yr OS: 0–28%).23 Majority of HCC patients present with VP 3 & 4, and as the mortality is high, usually surgery is not considered in these advanced cases. Among systemic therapies, sorafenib is an oral TKI that improves survival in patients with advanced HCC.14 However, SHARP trial demonstrated that tumour response to sorafenib treatment is only 2–3%.4 Subgroup analysis of patients with PVT showed that there is only marginal improvement of OS with sorafenib compared with placebo (median OS: 8.1 vs 4.9 month; P-value: NS).4,5 Severe adverse events are as high as 9–15% with sorafenib. Propensity-matched analysis also did not show any OS benefit with sorafenib in this patient cohort.7,14 Nakazawa et al. reported that RT significantly improves survival in patients with PVT compared with sorafenib (10.9 vs. 4.8 months, respectively; P = 0.025).7 In view of compromised blood flow and decompensated liver function status in PVT patients, TACE was not considered an ideal option in most of the situations apart from early PVTs (VP 1& 2). Median OS in PVT patients treated with TACE is between 5.6 and 8.7 months. Median OS in VP 1,2,3 and 4 PVT after TACE were 19.0, 11.0, 7.0 and 4.0 months, respectively. Interestingly, TACE responders (10.5 months) had significant survival benefit compared with non-responders (5.5 months).
TARE is a special type of trans-arterial treatment utilizing cytotoxic agent (131I or 90Y) in the hepatic artery.15,16 The effect of 131I in patients with PVT is controversial and still under investigation. 90Y, a β-emitting isotope, is the most popular radioembolization agent. TARE exhibits comparable efficacy with TACE for advanced HCC.16,17 TARE has a low risk of liver ischaemia because of the minimal embolic effects of the 90Y-glass microsphere. Therefore, TARE is more suitable in PVT. However, the response rate is only 28–50%, and median OS is poor (3–10 months).16,17 In fact, the recent network meta-analysis concluded that the combination of SBRT followed by TACE was associated with a higher survival than other combination modalities.
Radiation therapy was usually not a standard recommendation in HCC with vascular invasion. With rapid progress in RT delivery techniques with higher doses to portal vein without significant increase in radiation toxicity, radiation therapy is now being considered as an option.18 Total RT doses in literature vary from 17Gy to 60 Gy in 1.8–15 Gy per fraction, mostly because of the usage of different RT techniques. Response rates also vary between 27 and 54% with different response assessment criteria and total dose.23 Median OS is higher in responders (11–22 months) than in non-responders (5–7 months). Hence, response to treatment is an important predictor of outcome. Real-time image-guided radiation therapy techniques have the potential to deliver higher dose with minimal toxicities. There may be potential benefit with proton beam therapy and needs prospective evaluation. Proton therapy may work in different models in liver tumours as shown in recently published studies, and mean dose may not be an appropriate predictor of RILD as in photon therapy. There is a need for a further evaluation of toxicity and response to proton before routine clinical use.
Different treatment options and results are tabulated in Table 4. Few retrospective series have examined the role of SBRT in HCC with PVT, and median OS was 7–11 months (1-Yr: 30–40%, 2-Yr: 20–30%, 5-Yr OS: 5–20%). In a retrospective series by Nakazawa et al. comparing survival benefits of sorafenib vs RT in HCC with PVT (VP3 & 4), propensity-matched analysis (n = 28) showed better OS in the RT arm (10.9 months vs 4.8 months; P = 0.025) with no grade ≥3 GI or hepatic toxicity in the RT arm.7 Eun et al. compared RT with basic supportive care (BSC) in inoperable HCC (OS 45.9 Vs 4.8 months, P-value 0.001), and RT had significantly improved survival benefit. Cho et al. compared TACE plus RT Vs sorafenib in PVT (OS 10.9 Vs 4.8 months, P-value 0.002) and combined treatment was better than sorafenib alone.19 Li et al. compared PVT patients treated with and without RT after TACE, and RT group had better survival (10.9 Vs 4.1 months, P-value 0.001).20 Tang et al. compared surgery versus RT in PVT (1-yr OS: 42.2 Vs 51.6%, P-value 0.03).21 Shiozawa et al. compared SBRT with RFA in small (<5 cm) HCC, and the SBRT arm had better survival outcome (1-yr OS: 95.2% Vs 100%, P-value 0.08) (28). Sub-analyses of SHARP trial in BCLC C (n = 231) demonstrated that the sorafenib group (n = 108) achieved a longer OS (8.1 months vs 4.9 months) and time to progression (4.1 months vs 2.7 months) than the control (placebo) group (n = 123).22 Sapir et al. compared propensity-matched analysis in 206 HCC patients treated with TACE or SBRT. One- and 2-yr local control favoured SBRT: 97% and 91% for SBRT and 47% and 23% for TACE (Hazard ratio: 66.5, P < 0.001). Grade 3+ toxicity was 13% and 8% in TACE (P = 0.05) and RT respectively.18 Jung Ho IM et al. prospectively evaluated 985 HCC with PVT patients treated with SBRT.23 Half of the patients received RT to only PVT, and another half had RT to the entire lesion (liver mass and PVT). There was no difference in OS when PVT alone was irradiated in comparison when HCC with PVT are radiated together. However, toxicity was higher in patients who received RT to the entire lesion. Matsuo et al. compared different SBRT techniques (CK, arc-based treatment, 3DCRT) in a small cohort, and the CK group had superiority in both dosimetric parameters and response to treatment.24,25 There is no randomized study comparing outcomes between TARE and SBRT.26,27 Toxicity, cost-effectiveness and patient selection criteria are different in TARE and SBRT. Patients with diffuse liver lesion, with no pulmonary shunt and without portal vein thrombosis are indications for TARE.15,17 In portal vein thrombosis, SBRT for the portal vein thrombosis to achieve recanalization followed by TARE for diffuse lesion improves efficacy and reduces toxicity. Interim analysis data from the present study along with other published literature have helped in including SBRT for portal vein thrombosis as a treatment option in consensus guideline.25
In our study with a mean follow-up of 12.2 months, more than 6 and 12 months’ OS was 59% and 23%, respectively. Recanalization occurred in 82 of 139 (59%) patients, and in patients with response (recanalization), there was significant improvement in OS. Mean OS in responders was 18.4 months and only 9.34 months in non-responders. Six- and 12-month survival in responders was 74% and 48%, whereas in non-responders, it was 66% and 24.5%, respectively. Hence, two-third of the patients (responders) who underwent treatment had the potential of having survival advantage after treatment. Only 5 patients (4%) with recanalization had further treatment with TARE or TACE, but still there was significant survival advantage in patients with recanalization, suggesting that recanalization prevents or delays decompensation of liver function and helps in improvement of survival function. About 13% of the patients had unplanned admission mostly due to ascites and decompensation (7%). Ten (7%) patients had ‘classic RILD’, and 26 (19%) patients had ‘non-classic’ RILD. More than 2-point decrease in the CP score was observed in 24 (20%) patients.
Higher dose was expected to have better response, but in our study, response rate was not influenced by the RT dose after a threshold dose. There was interesting observation in patients eligible for treatment but who had not given consent and were treated with conventional treatment (TKIs) or best supportive care. OS in eligible but not treated (non-accrual) patients (n = 16) was 5.6 months, and that in patients not eligible (poor PS, ascites) (n = 23) was 2.6 months only. There was significant survival advantage in patients treated with SBRT and had ‘response to treatment’ compared with eligible patients treated with TKIs. Randomized study in portal vein thrombosis treated with SBRT and TKIs comparing with TKIs alone may be ideal. However, with the present survival advantage with SBRT for portal vein thrombosis, it may not be ethical to not offer this treatment to suitable patients and hence unlikely to have a randomized study addressing this issue. In this scenario, the present adequately powered prospective study provides a reasonable evidence for recommendation to treat major portal vein thrombosis (VP3/4) with SBRT. The present study does have limitations in terms of short follow-up (13 months median follow-up), and hence long-term complications are not reported. The present study is a phase II prospective study, and there is a need for a randomized controlled study with or without SBRT to have level I evidence regarding the efficacy of SBRT. Radiomics and molecular markers are required for further prognostication.
In summary, the present appropriately powered prospective study with relative long-term follow-up showed significant survival advantage among patients with ‘response to treatment’ after SBRT to the PVT. There was doubling of mean OS in responders after SBRT compared with ‘non-responders’. No patient- or treatment-related factors influenced ‘response to treatment’. It seems response to treatment alone is the only factor that determines improvement in survival function. In future, molecular and radiomic studies to prognosticate response to treatment need to be studied. CK-based SBRT is a safe and effective treatment option for HCC with PVT. For HCC with PVT patients for whom curative local treatment options are limited, SBRT may lead to a sustained local control, survival advantage with a low risk of complications in responders. SBRT for PVT is included as an option in consensus guidelines. However, this small but significant survival benefit among responders needs to be further studied for cost-effectiveness and clinical relevance in different ethnic groups.
CREDIT AUTHORSHIP CONTRIBUTION STATEMENT
Conceptualization: Dutta D, Sudhindran S.
Data curation: Dutta D, Sreenija Y, Sruthi K.
Formal analysis: Dutta D, Sreenija Y.
Investigation: Nair H, Sashidharan A, Nimmya K, Kannan R, George S, Edappattu A MSc, Haridas NK, Jose WM, Keechilat P, Valsan A, Koshy A, Gopalakrishna R, Sadasivan S, Gopalakrishnan U, Balakrishnan D, Sudheer OV.
Methodology: Dutta D, Sreejina Y, Sruthi K.
Project administration: Sreenija Y.
Software: Sreenija Y, Nimmya K.
Supervision: Dutta D.
Validation: Dutta D, Keechilat P, Sudhindran S.
Visualization: Dutta D, Sreejina Y, Sruthi K.
Writing - original draft: Dutta D, Sreenija Y.
Writing - review & editing: Dutta D, Sreenija Y, Sruthi K, Nair H, Sudhindran S, Keechilat P, Sudheer OV, Sadasivan S.
Conflicts of interest
All authors have none to declare.
Acknowledgements
We thank Amrita Institute of Medical Science, Kochi, Kerala, India for seed grant support.
Funding
None.
Study presented
This study was presented as discussion at the European Society of Therapeutics and Radiation therapy (ESTRO) Annual conference 2023 [Radiat Oncol 2023; DOI: 10.1016/S0167-8140(23)09020-5].
Disclaimer
None.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jceh.2024.101404.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Fig. SUPPL 1.
Overall survival of entire cohort (n = 139)
Fig. SUPPL 2.
Overall survival with different performance status patients
References
- 1.Siegel R., Ma J., Zou Z., Jemal A. Cancer statistics, 2014. CA Cancer J Clin. 2014;64:9–29. doi: 10.3322/caac.21208. [DOI] [PubMed] [Google Scholar]
- 2.Llovet J.M., Burroughs A., Bruix J. Hepatocellular carcinoma. Lancet Lond Engl. 2003 Dec 6;362:1907–1917. doi: 10.1016/S0140-6736(03)14964-1. [DOI] [PubMed] [Google Scholar]
- 3.Forner A., Ayuso C., Varela M., et al. Evaluation of tumor response after locoregional therapies in hepatocellular carcinoma. Cancer. 2009;115:616–623. doi: 10.1002/cncr.24050. [DOI] [PubMed] [Google Scholar]
- 4.Rimassa L., Santoro A. Sorafenib therapy in advanced hepatocellular carcinoma: the SHARP trial. Expert Rev Anticancer Ther. 2009;9:739–745. doi: 10.1586/era.09.41. [DOI] [PubMed] [Google Scholar]
- 5.Cheng A.L., Kang Y.K., Chen Z., et al. Efficacy and safety of sorafenib in patients in the Asia-Pacific region with advanced hepatocellular carcinoma: a phase III randomised, double-blind, placebo-controlled trial. Lancet Oncol. 2009;10:25–34. doi: 10.1016/S1470-2045(08)70285-7. [DOI] [PubMed] [Google Scholar]
- 6.Colagrande S., Inghilesi A.L., Aburas S., Taliani G.G., Nardi C., Marra F. Challenges of advanced hepatocellular carcinoma. World J Gastroenterol. 2016;22:7645. doi: 10.3748/wjg.v22.i34.7645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nakazawa T., Hidaka H., Shibuya A., et al. Overall survival in response to sorafenib versus radiotherapy in unresectable hepatocellular carcinoma with major portal vein tumor thrombosis: propensity score analysis. BMC Gastroenterol. 2014 May 3;14:84. doi: 10.1186/1471-230X-14-84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Okuda K., Ohtsuki T., Obata H., et al. Natural history of hepatocellular carcinoma and prognosis in relation to treatment study of 850 patients. Cancer. 1985;56:918–928. doi: 10.1002/1097-0142(19850815)56:4<918::aid-cncr2820560437>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
- 9.Dutta D., Tatineni T., Yarlagadda S., et al. Hepatocellular carcinoma patients with portal vein thrombosis treated with robotic radiosurgery: interim results of a prospective study. Indian J Gastroenterol Off J Indian Soc Gastroenterol. 2021 Aug;40:389–401. doi: 10.1007/s12664-021-01172-w. [DOI] [PubMed] [Google Scholar]
- 10.Dutta D., Kataki K.J., George S., et al. Prospective evaluation of fiducial marker placement quality and toxicity in liver CyberKnife stereotactic body radiotherapy. Radiat Oncol J. 2020 Nov 30;38:253–261. doi: 10.3857/roj.2020.00472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Dawson L.A., Normolle D., Balter J.M., McGinn C.J., Lawrence T.S., Ten Haken R.K. Analysis of radiation-induced liver disease using the Lyman NTCP model. Int J Radiat Oncol. 2002 Jul 15;53:810–821. doi: 10.1016/s0360-3016(02)02846-8. [DOI] [PubMed] [Google Scholar]
- 12.Sun J.X., Shi J., Li N., et al. Portal vein tumor thrombus is a bottleneck in the treatment of hepatocellular carcinoma. Cancer Biol Med. 2016 Dec;13:452–458. doi: 10.20892/j.issn.2095-3941.2016.0059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yoon S.M., Lim Y.S., Won H.J., et al. Radiotherapy plus transarterial chemoembolization for hepatocellular carcinoma invading the portal vein: long-term patient outcomes. Int J Radiat Oncol. 2012 Apr 1;82:2004–2011. doi: 10.1016/j.ijrobp.2011.03.019. [DOI] [PubMed] [Google Scholar]
- 14.Bruix J., Cheng A.L., Meinhardt G., Nakajima K., De Sanctis Y., Llovet J. Prognostic factors and predictors of sorafenib benefit in patients with hepatocellular carcinoma: analysis of two phase III studies. J Hepatol. 2017 Nov 1;67:999–1008. doi: 10.1016/j.jhep.2017.06.026. [DOI] [PubMed] [Google Scholar]
- 15.Vilgrain V., Pereira H., Assenat E., et al. Efficacy and safety of selective internal radiotherapy with yttrium-90 resin microspheres compared with sorafenib in locally advanced and inoperable hepatocellular carcinoma (SARAH): an open-label randomised controlled phase 3 trial. Lancet Oncol. 2017 Dec 1;18:1624–1636. doi: 10.1016/S1470-2045(17)30683-6. [DOI] [PubMed] [Google Scholar]
- 16.Lau W.Y., Sangro B., Chen P.J., et al. Treatment for hepatocellular carcinoma with portal vein tumor thrombosis: the emerging role for radioembolization using yttrium-90. Oncology. 2013;84:311–318. doi: 10.1159/000348325. [DOI] [PubMed] [Google Scholar]
- 17.She W.H., Cheung T.T., Yau T.C.C., et al. Survival analysis of transarterial radioembolization with yttrium-90 for hepatocellular carcinoma patients with HBV infection. Hepatobiliary Surg Nutr. 2014 Aug;3:185–193. doi: 10.3978/j.issn.2304-3881.2014.07.09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sapir E., Tao Y., Schipper M.J., et al. Stereotactic body radiation therapy as an alternative to transarterial chemoembolization for hepatocellular carcinoma. Int J Radiat Oncol. 2018 Jan 1;100:122–130. doi: 10.1016/j.ijrobp.2017.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cerrito L., Annicchiarico B.E., Iezzi R., Gasbarrini A., Pompili M., Ponziani F.R. Treatment of hepatocellular carcinoma in patients with portal vein tumor thrombosis: beyond the known frontiers. World J Gastroenterol. 2019 Aug 21;25:4360–4382. doi: 10.3748/wjg.v25.i31.4360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kang J., Nie Q., Du R., et al. Stereotactic body radiotherapy combined with transarterial chemoembolization for hepatocellular carcinoma with portal vein tumor thrombosis. Mol Clin Oncol. 2014 Jan 1;2:43–50. doi: 10.3892/mco.2013.196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Tang Q he, Li A jun, Yang G ming, et al. Surgical resection versus conformal radiotherapy combined with TACE for resectable hepatocellular carcinoma with portal vein tumor thrombus: a comparative study. World J Surg. 2013 Jun 1;37:1362–1370. doi: 10.1007/s00268-013-1969-x. [DOI] [PubMed] [Google Scholar]
- 22.Kim P.H., Choi S.H., Kim J.H., Park S.H. Comparison of radioembolization and sorafenib for the treatment of hepatocellular carcinoma with portal vein tumor thrombosis: a systematic review and meta-analysis of safety and efficacy. Korean J Radiol. 2019 Jan 30;20:385–398. doi: 10.3348/kjr.2018.0496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Im J.H., Yoon S.M., Park H.C., et al. Radiotherapeutic strategies for hepatocellular carcinoma with portal vein tumour thrombosis in a hepatitis B endemic area. Liver Int. 2017;37:90–100. doi: 10.1111/liv.13191. [DOI] [PubMed] [Google Scholar]
- 24.Matsuo Y., Yoshida K., Nishimura H., et al. Efficacy of stereotactic body radiotherapy for hepatocellular carcinoma with portal vein tumor thrombosis/inferior vena cava tumor thrombosis: evaluation by comparison with conventional three-dimensional conformal radiotherapy. J Radiat Res. 2016 Sep 1;57:512–523. doi: 10.1093/jrr/rrw028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kumar A., Acharya S.K., Singh S.P., et al. 2023 update of Indian national association for study of the liver consensus on management of intermediate and advanced hepatocellular carcinoma: the Puri III recommendations. J Clin Exp Hepatol. 2023 Mar-Apr;13:273–302. doi: 10.1016/j.jceh.2023.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Srivastava A., Parambath H.K., Ramdulari A.V., et al. Is hepatocellular carcinoma complicated with portal vein tumor thrombosis potentially curable by radiotherapy in the form of stereotactic body radiation therapy? Int J Radiat Biol. 2022;98:1495–1509. doi: 10.1080/09553002.2022.2055800. [DOI] [PubMed] [Google Scholar]
- 27.Rim C.H., Kim C.Y., Yang D.S., Yoon W.S. Comparison of radiation therapy modalities for hepatocellular carcinoma with portal vein thrombosis: a meta-analysis and systematic review. Radiother Oncol. 2018 Oct;129:112–122. doi: 10.1016/j.radonc.2017.11.013. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.