Patients and treatments
A total of 226 patients with HCC were eligible and enrolled in the present study including 210 patients initially received TACE and the remaining 16 were treated with SBRT-IO. Table 1 shows the baseline and tumour characteristics of all the patients and their significances for clinical outcome.
Table 1
Baseline demographics and tumour characteristics of all patients
| Before propensity score matching | After propensity score matching |
| Unmatched TACE N = 210 | SBRT-IO N = 16 | P value | Matched TACE N = 48 | P value |
Age (median, range) years | 69 (36–94) | 66.5 (38–86) | 0.504 | 73 (49–87) | 0.149 |
Sex (n, % male) | 158 (75.2) | 14 (87.5) | 0.268 | 43 (89.6) | 0.817 |
Hepatitis B carrier (n, %) | 129 (61.4) | 12 (75.0) | 0.280 | 26 (54.2) | 0.142 |
ECOG 0–1 (n, %) | 192 (91.4) | 12 (75.0) | 0.439 | 45 (93.8) | 0.407 |
Child-Pugh class A (n, %) | 182 (86.7) | 14 (87.5) | 0.925 | 46 (95.8) | 0.233 |
ALBI grade 1 2 3 | 76 (36.2) 121 (57.6) 13 (6.2) | 8 (50.0) 7 (43.8) 1 (6.2) | 0.531 | 15 (31.2) 32 (66.7) 1 (2.1) | 0.238 |
Albumin (g/L) | 37 (17–48) | 39 (30–45) | 0.250 | 37 (25–45) | 0.192 |
Bilirubin (µmol/L) | 13 (4–55) | 15 (8–122) | 0.171 | 12.5 (4–39) | 0.149 |
Platelet (x109/L) | 169.5 (25–551) | 234 (79–402) | 0.069 | 226 (66–522) | 0.773 |
INR | 1.1 (0.8–2.3) | 1.1 (1–1.5) | 0.339 | 1.1 (0.9–1.6) | 0.543 |
BCLC stage (n, %) A B C | 79 (37.6) 99 (47.2) 32 (15.2) | 3 (18.8) 5 (33.3) 8 (50) | 0.002 | 9 (18.7) 15 (31.3) 24 (50) | 0.998 |
Tumour number (n, %) 1 2 ≥ 3 | 89 (42.4) 26 (12.4) 95 (45.2) | 9 (56.2) 2 (12.5) 5 (31.3) | 0.518 | 27 (56.3) 1 (2.1) 20 (41.6) | 0.460 |
Tumour size (cm)* | 6.95 (1–19.6) | 10 (3.4–18) | 0.016 | 10.4 (2.68–19.6) | 1.000 |
Portal vein invasion (n, %) | 19 (9.1) | 3 (18.8) | 0.001 | 10 (20.8) | 0.827 |
AFP ≥ 200 ng/ml (n, %) Range | 84 (40.0) 1-1458960 | 7 (43.8) 3-499988 | 0.768 | 21 (43.8) 2-362901 | 1.000 |
The SBRT-IO group had higher percentage of patients with Barcelona Clinic Liver Cancer (BCLC) stage C disease and portal vein invasion; median size of tumour was larger in the SBRT-IO group (10 cm vs. 6.95 cm, p = 0.016). After propensity score matching, a total of 48 patients treated with TACE were identified to match the 16 patients treated with SBRT-IO. No significant difference was observed between-group. Overall, around 90% of analysed patients were male and had performance status of ECOG 0–1, and 60% were hepatitis B carrier.
Patients Population Between Tace And Sbrt-io
Among the 64 included patients after matching, the median size of tumour was 10 cm (range: 3.4–19.6 cm) and 20.3% of the patients had portal vein invasion. Among the 48 patients matched in the TACE arm, median 2 sessions of TACE (range: 1–16) were given. For the SBRT-IO arm, a median dose of 35 Gy (range: 27.5–37.5 Gy) was prescribed and median 10 cycles of nivolumab (range: 1–20 doses) were given. Total 24 lesions were irradiated in SBRT-IO arm. (N = 11, single lesion; N = 2, two lesions; N = 3, three lesions). One patient and six patients received post-progression therapies in the SBRT-IO and TACE arms, respectively. Neither patient in the TACE arm received SBRT or immunotherapy, nor patient in SBRT-IO arm received TACE after progression (refer to Table S1 for detailed information).
Overall Survival And Progression-free Survival Between Tace And Sbrt-io
The survival data was censored on December 31st, 2020. The median follow-up time of the SBRT-IO and matched TACE groups were 12.7 months (range: 2.5–36.1 months) and 7.4 months (range: 0.2–57.2 months), respectively. The 6-month, 12-month, and 24-month PFS were better in the SBRT-IO group (93.3% vs. 37.5%, 93.3% vs. 16.7%, and 77.8% vs. 2.1%, respectively, p < 0.001). The median PFS of the SBRT-IO group was not reached (range: 1.9–36.1 months) compared to 4.83 months (range: 0.2–42.2 months) of the TACE group. The 6-month, 12-month, and 24-month OS were also better in the SBRT-IO group (93.8% vs. 54.2%, 93.8% vs. 31.3%, and 80.4% vs. 8.3%, respectively, p < 0.001). The median OS of the SBRT-IO group was not reached (range: 2.5–36.1 months) compared to 7.44 months (range: 0.2–57.2 months) of the TACE group (as shown in Fig. 1).
At the time of analysis, there were three deaths out of 16 patients in the SBRT-IO arm, and all the 48 patients died in the matched TACE group. All the three patients that died in the SBRT-IO group showed no evidence of disease progression; two died of community-acquired pneumonia and one died of haemobilia. In the matched TACE group, 37 deaths (77.1%) were cancer related, four died of pulmonary causes, two died of intra-cranial haemorrhage, and one died of myocardial infarction, liver abscess, liver decompensation, intestinal obstruction, or unknown cause (each). Among patients in the matched TACE group, intra-hepatic progression (90.6%) represented the dominant mode of failure; there were no significant differences in the PFS or OS between the different treatment periods from 2010–2020 (as shown in Fig. S1).
Under the multi-variable analysis, SBRT-IO, was an independent predictor of better OS (hazard ratio [HR] = 0.14, range: 0.30–0.96, p = 0.036) and better PFS (HR = 0.1, range: 0.03–0.33, p < 0.001). Tumour number was another independent predictor of OS (HR = 0.54, range: 0.30–0.96, p = 0.036) and PFS (HR = 0.38, range 0.21–0.72, p = 0.003) (Table 2).
Table 2
Univariate and multivariate analyses of potential prognostic factors affecting overall and progression-free survival after propensity score matching
| Overall Survival | Progression-free Survival |
| UVA | MVA | UVA | MVA |
| HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P | HR | 95% CI | P |
For matched groups (n = 64): | | | | | | | | | | | | |
SBRT-IO vs. TACE | 0.13 | 0.04–0.42 | < 0.001 | 0.14 | 0.04–0.46 | 0.001 | 0.10 | 0.03–0.32 | < 0.001 | 0.10 | 0.03–0.33 | < 0.001 |
Age (< 60 vs. ≥60 years) | 1.24 | 0.58–2.67 | 0.58 | | | | 1.37 | 0.64–2.94 | 0.42 | | | |
Sex (male vs. female) | 1.25 | 0.49–3.16 | 0.64 | | | | 1.50 | 0.59–3.81 | 0.39 | | | |
Hepatitis B carrier (yes vs. no) | 1.03 | 0.59–1.81 | 0.92 | | | | 0.99 | 0.56–1.73 | 0.97 | | | |
ECOG (0–1 vs. 2) | 1.92 | 0.75–4.90 | 0.17 | | | | 2.27 | 0.89–5.81 | 0.09 | | | |
Child-Pugh class (A vs. B) | 2.08 | 0.50–8.61 | 0.31 | | | | 1.91 | 0.46–7.92 | 0.37 | | | |
ALBI grade (1 vs. 2) | 0.63 | 0.34–1.15 | 0.13 | | | | 0.52 | 0.28–0.96 | 0.04 | 0.90 | 0.49–1.66 | 0.73 |
Portal vein invasion (Yes vs. no) | 1.08 | 0.55–2.12 | 0.82 | | | | 1.13 | 0.51–2.50 | 0.76 | | | |
BCLC stage (A vs. B) BCLC stage (A vs. C) | 1.15 0.77 | 0.50–2.64 0.36–1.68 | 0.74 0.52 | | | | 1.19 0.77 | 0.46–3.08 0.31–1.87 | 0.72 0.56 | | | |
Tumour number (n = 1–2 vs. ≥ 3) | 0.45 | 0.25–0.80 | 0.007 | 0.54 | 0.30–0.96 | 0.036 | 0.38 | 0.21–0.70 | 0.002 | 0.38 | 0.21–0.72 | 0.003 |
Tumour size (< 10 cm vs. ≥ 10 cm) | 0.60 | 0.34–1.08 | 0.09 | | | | 0.59 | 0.33–1.05 | 0.07 | | | |
AFP (< 200 vs. ≥ 200 ng/ml) | 0.71 | 0.40–1.25 | 0.24 | | | | 0.67 | 0.38–1.19 | 0.17 | | | |
Abbreviations: TACE, transarterial chemoembolisation; SBRT-IO, combined stereotactic body radiotherapy and immunotherapy; ECOG, Eastern Cooperative Oncology Group; INR, international normalised ratio; BCLC, Barcelona Clinic Liver Cancer; AFP, alpha-feto protein; UVA: univariate analysis; MVA: multivariate analysis; HR, hazard ratio; CI, confidence interval |
Radiological Objective Response Rate Between Tace And Sbrt-io
Figure 2 depicts the best objective response of 16 patients in SBRT-IO arm and 42 evaluable patients in matched-TACE arm. Six patients of TACE arm did not have reassessment scan due to rapid deterioration. The ORR was significantly higher in the SBRT-IO group (87.5% vs. 16.7%, p < 0.001). DCR was also significantly better in the SBRT-IO group (81.3% vs. 37.5%, p = 0.002). In the matched TACE group, 24 patients (50%) never had radiological disease controlled. In contrast, only one patient (6.3%) developed progressive disease after SBRT-IO; this patient developed a new HCC focus outside the irradiated field and two SBRT-treated lesions had partial response (PR) and static disease (SD). Two patients of the SBRT-IO arm (12.5%) had radiofrequency ablation (RFA) performed after PR, with complete clearance of the tumour subsequently achieved. The waterfall plot of Fig. 3 illustrates the better tumour shrinkage of index lesion in the SBRT-IO and TACE group.
Of note, Nivolumab was stopped for eight patients who achieved CR after median 7.1 months of treatment (range: 2.1–15.6 months); none of them developed relapse in the median follow-up time of 5.7 months (range: 0.7–25.0 months).
Treatment related adverse events and liver function deterioration between TACE and SBRT-IO
Risk of ≥ grade 3 TRAEs and discontinuation of treatment due to toxicities were more common in patients who received TACE (60.4% vs. 18.8%, p = 0.004; 25% vs. 12.5%, respectively, p = 0.295). There was more elevated transaminase, anaemia, leukopenia, and fever in TACE group, while patients who received SBRT-IO had more fatigue, diarrhoea, and rash. Among patients treated with SBRT-IO, none developed classical radiation induced liver disease, and there were no treatment-related deaths reported. There were fewer patients who developed Child-Pugh score progression ≥ 2 at 3 months (6.7% vs. 20.9%, p = 0.008), 6 months (6.7% vs. 12.0%, p = 0.021), and 12 months (0% vs. 21.4%, p < 0.001) in the SBRT-IO arm compared to the TACE arm (Table 3).
Table 3
Treatment related adverse event and Child-Pugh score progression of SBRT-IO vs. matched TACE
| SBRT-IO (N = 16) | Matched TACE (N = 48) | |
| Any Grade | Grade 3–4 | Any Grade | Grade 3–4 | P value |
| Number (%) | Number (%) | |
Treatment-related AEs | | 3 (18.8%) | | 29 (60.4%) | 0.004 |
AEs lead to discontinuation | 2 (12.5%) | 1 (6.3%) | 12 (25%) | 5 (10.4%) | 0.295 |
Treatment-related death | 0 (0%) | 0 (0%) | 2 (4.2%) | 1 (2.1%) | 0.407 |
Haemoglobin | 8 (50%) | 0 (0%) | 38 (79.2%) | 4 (8.3%) | 0.06 |
Leukocytes | 2 (12.5%) | 0 (0%) | 24 (50%) | 0 (0%) | 0.025 |
Platelet | 12 (75%) | 1 (6.3%) | 26 (54.2%) | 6 (12.5%) | 0.251 |
Bilirubin | 5 (31.3%) | 1 (6.3%) | 20 (41.7%) | 5 (10.4%) | 0.617 |
AST/ALT | 15 (93.7%) | 0 (0%) | 43 (89.6%) | 25 (52.1%) | < 0.001 |
Nausea and vomiting | 4 (25%) | 0 (0%) | 13 (27.1%) | 0 (0%) | 0.456 |
Diarrhoea | 3 (18.8%) | 0 (0%) | 1 (2%) | 0 (%) | 0.002 |
Appetite lost | 1 (6.3%) | 0 (0%) | 9 (18.9%) | 0 (0%) | 0.477 |
Fatigue | 10 (62.5%) | 0 (0%) | 9 (18.7%) | 2 (4.2%) | 0.003 |
Fever | 3 (18.8%) | 0 (0%) | 23 (47.9%) | 0 (0%) | 0.04 |
Weight loss | 1 (6.3%) | 0 (0%) | 8 (16.7%) | 0 (0%) | 0.562 |
Pain | 3 (18.8%) | 0 (0%) | 24 (50%) | 0 (0%) | 0.028 |
Rash | 5 (31.3%) | 0 (0%) | 4 (8.3%) | 0 (0%) | 0.002 |
Pruritus | 2 (12.5%) | 1 (6.3%) | 4 (8.3%) | 0 (0%) | 0.213 |
Adrenal insufficiency | 1 (6.3%) | 0 (0%) | 1 (2.1%) | 0 (0%) | 0.407 |
Progression of CP score ≥ 2 | | | | | |
3 months | 1/15 (6.7%) | 9/43 (20.9%) | 0.008 |
6 months | 1/15 (6.7%) | 3/25 (12.0%) | 0.021 |
12 months | 0/8 (0%) | 3/14 (21.4%) | < 0.001 |
Abbreviations: TACE, transarterial chemoembolisation; SBRT-IO, combined stereotactic body radiotherapy and immunotherapy; AEs, adverse events; AST: Aspartate transaminase; ALT: Alanine transaminase; CP, Child-Pugh |
* The incidence of only toxicities ≥ 5% is shown |