Twelve patients were enrolled in this study. All patients are shown in Table 1.
Case | Age | Sex | BCLC | Treatment line | Previous treatment | mALBI Grade | Child–Pugh | Etiology | Time from initial TKI (months) |
1 | 75 | M | C | 4 | Reg | G2a | 5 | NASH | 26 |
2 | 86 | M | B | 2 | Len (with -drawal) | G2b | 6 | NASH | 16 |
3 | 64 | M | C | 6 | Len | G2b | 7 | NASH | 24 |
4 | 74 | M | C | 5 | Ram | G2b | 6 | NASH | 22 |
5 | 66 | F | C | 3 | Len | G2b | 6 | C | 38 |
6 | 45 | M | C | 4 | Reg | G1 | 5 | B | 19 |
7 | 76 | F | C | 3 | Ram | G2a | 5 | NASH | 15 |
8 | 66 | M | B | 4 | Len | G2b | 6 | B | 39 |
9 | 67 | F | B | 2 | Len | G3 | 8 | B | 9 |
10 | 80 | M | B | 4 | Reg | G2b | 6 | C | 19 |
11 | 71 | M | C | 2 | Len | G2a | 5 | ASH | 10 |
12 | 73 | M | C | 3 | Sor | G2a | 5 | ASH | 12 |
The Child–Pugh score was Child–Pugh A in all patients in IMbrave150, but two patients with Child–Pugh B (7 and 8 points) were included in this study, reflecting real clinical practice. The median treatment line was 3.5 and ranged from second line to sixth line. The tumor status was BCLC B in four cases and BCLC C in eight cases. There was no significant change in ALBI grade during the treatment period from the start to the third course (week 9), and there was no deterioration in ALBI grade (Fig. 1).
Tumor size changes are shown in spider plots in Fig. 2a,2b.
One out of 12 patients had a partial response, two had no change, and four had tumor growth. Three patients had a partial response, six had stable disease, and three had progressive disease according to mRECIST (Table 2).
Cases | n = 12 |
Age | 72 (45–86) |
Gender (M / F) | 10 (83%) / 2 (17%) |
BCLC B / C | 4 (33%) / 8 (66%) |
up-to7 in /out | 1 (25%) / 3 (75%) |
Treatment line (2/3/4/5/6) | 3 (25%) / 3 (25%) / 4 (33%) / 1 (8%) / 1 (8%) |
mALBI grade (1/2a/2b/3) | 1 (8%) / 4 (33%) / 6 (50%) / 1 (8%) |
Child –Pugh (5/6/7/8) (score) | 5 (42%) / 5 (42%) / 1 (8%) / 1 (8%) |
Etiology (HBV HCV/NASH/ASH) | 3 (25%) / 2 (17%) / 5 (42%) / 2 (17%) |
Time from initial TKI (Month) | 19 (9–39) |
Among the eight cases in which we were able to measure the pretreatment growth rate and compare it with the post-treatment rate, one case had a tumor disappearance rate greater than or equal to two, which met the definition of hyperprogression [16]. In this case, the tumor markers decreased during the treatment, and the tumor began to shrink after further treatment (case 8; Fig.3a).
The change in tumor size in the early phase after the start of treatment varied considerably from case to case. In one case, there was a marked reduction in tumor size and the appearance of ascites immediately after the first course of treatment, but the tumor grew again immediately after the second course of treatment (Fig.3b).
The median PFS was 2.7 months, which was shorter than that in the IMbrave150 trial. In one case, the tumor was extremely large (28 cm) from the outset and rapidly grew before treatment, rupturing on day 7 of treatment (case 9). Analysis of spider plots by pretreatment showed that the treatment effect of atezolizumab plus bevacizumab tended to differ for each pretreatment (Fig.4a.b.c.d).
In the case of atezolizumab plus bevacizumab after treatment with lenvatinib, there was a rapid increase in tumor size followed by a decrease. However, for atezolizumab plus bevacizumab after treatment with regorafenib and sorafenib, all cases were almost SD. In patients with atezolizumab plus bevacizumab after treatment with ramucirumab, there was a rapid decrease in tumor size followed by an increase.
Treatment-related side effects were observed in many patients (Table 3).
n(%) | IMbrave150 (n=329) | LINKS (n=12) |
Any grade | Grade 3 or 4 | Any grade | Grade 3 or 4 |
Hypertension | 98 (29.8) | 50 (15.2) | 1 (8.3) | 0 |
Fatigue | 67 (20.4) | 8 (2.4) | 0 | 0 |
Proteinuria | 66 (20.1) | 10 (3.0) | 4 (33.3) | 3 (25) |
AST increase | 64 (19.5) | 23(7.0) | 2 (16.6) | 2 (16.6) |
Pruritus | 64 (19.5) | 0 | 0 | |
Diarrhea | 62 (18.8) | 6 (1.8) | 0 | 0 |
Decreased appetite | 58 (17.6) | 4 (1.2) | 0 | 0 |
Pyrexia | 59 (17.9) | 4 (1.2) | 2 (16.6) | 0 |
A༬T increase | 46 (16.0) | 12 (3.6) | 2 (16.6) | 0 |
Constipation | 44 (13.4) | 0 | 0 | |
Blood bilirubin increase | 43 (13.1) | 8 (2.4) | 4 (33.3) | 2 (16.6) |
Rash | 41 (12.5) | 0 | 0 | |
Abdominal pain | 40 (12.2) | 4 (1.2) | 1 (8.3) | 0 |
Weight decrease | 37 (11.2) | 8 (2.4) | 0 | |
Asthenia | 22 (6.7) | 1 (0.3) | 1 (8.3) | 0 |
Infusion reaction | 37 (11.2) | 8 (2.4) | 0 | 0 |
HFS | 3 (0.9) | 0 | 1 (8.3) | 0 |
HCC Rupture | - | - | 2 (16.6) | 2 (16.6) |
Hepatic encephalopathy | - | - | 1 (8.3) | 1 (8.3) |
Ascites | - | - | 2 (16.6) | 2 (16.6) |
Steven Johnson Syndrome | - | - | 1 (8.3) | 1 (8.3) |
Compared with the IMbrave150 study without prior treatment, there were fewer side effects commonly seen with TKIs in general, such as fatigue, hypertension, anorexia, and abdominal pain. Conversely, the most common side effects, or those that were not observed in IMbrave150 but appeared in this study, were symptoms associated with liver function and liver cancer progression, such as ascites, encephalopathy, liver cancer rupture, and elevated bilirubin. Two patients who had large tumors before treatment initiation had HCC rupture immediately after treatment and at 5.5 months. Five out of 12 patients had their treatment discontinued. The reasons for discontinuation were progressive disease in one case, liver cancer rupture in two cases, ascites jaundice in one case, and Stevens–Johnson syndrome in one case.