Boosted selective internal radiation therapy with 90Y-loaded glass microspheres (B-SIRT) for hepatocellular carcinoma patients: a new personalized promising concept

Abstract

Purpose: To evaluate the impact of dosimetry based on MAA SPECT/CT for the prediction of response, toxicity and survival, and for treatment planning in patients with hepatocellular carcinoma (HCC) treated with (90)Y-loaded glass microspheres (TheraSphere®).

Methods: TheraSphere® was administered to 71 patients with inoperable HCC. MAA SPECT/CT quantitative analysis was used for the calculation of the tumour dose (TD), healthy injected liver dose (HILD), and total injected liver dose. Response was evaluated at 3 months using EASL criteria. Time to progression (TTP) and overall survival (OS) were evaluated using the Kaplan-Meier method. Factors potentially associated with liver toxicity were combined to construct a liver toxicity score (LTS).

Results: The response rate was 78.8%. Median TD were 342 Gy for responding lesions and 191 Gy for nonresponding lesions (p < 0.001). With a threshold TD of 205 Gy, MAA SPECT/CT predicted response with a sensitivity of 100% and overall accuracy of 90%. Based on TD and HILD, 17 patients underwent treatment intensification resulting in a good response rate (76.4%), without increased grade III liver toxicity. The median TTP and OS were 5.5 months (2-9.5 months) and 11.5 months (2-31 months), respectively, in patients with TD <205 Gy and 13 months (10-16 months) and 23.2 months (17.5-28.5 months), respectively, in those with TD >205 Gy (p = 0.0015 and not significant). Among patients with portal vein thrombosis (PVT) (n = 33), the median TTP and OS were 4.5 months (2-7 months) and 5 months (2-8 months), respectively, in patients with TD <205 Gy and 10 months (6-15.2 months) and 21.5 months (12-28.5 months), respectively, in those with TD >205 Gy (p = 0.039 and 0.005). The median OS was 24.5 months (18-28.5 months) in PVT patients with TD >205 Gy and good PVT targeting on MAA SPECT/CT. The LTS was able to detect severe liver toxicity (n = 6) with a sensitivity of 83% and overall accuracy of 97%.

Conclusion: Dosimetry based on MAA SPECT/CT was able to accurately predict response and survival in patients treated with glass microspheres. This method can be used to adapt the injected activity without increasing liver toxicity, thus defining a new concept of boosted selective internal radiation therapy (B-SIRT). This new concept and LTS enable fully personalized treatment planning with glass microspheres to be achieved.

Fig. 1

VOI delineation in a 77-year-old patient with a large HCC of the right lobe associated with two small lesions of the left lobe who underwent treatment intensification (boosted). a Initial CT slice shows heterogeneous HCC of 12.3 cm with a central area of necrosis. b, d MAA SPECT/CT with high uptake and a hypofixing area: b VOI delineation for the tumour; d VOI delineation for the injected liver. The patient underwent treatment intensification with 3.5 GBq of ⁹⁰Y-loaded glass microspheres (ILD 162 Gy, TD 280 Gy, HILD 54 Gy). Using the standard approach, only 2.6 GBq of ⁹⁰Y-loaded glass microspheres would have been used to achieve an ILD of 120 Gy, then the TD would have been only 207 Gy, and the HILD 38 Gy. c CT slice 3 months after injection shows EASL partial response of the treated tumour of the right lobe. The left lesions were treated by hyperselective chemoembolization due to their small size. The patient was still alive, but with progressive disease, at 28.5 months (most recent follow-up visit)

Fig. 2

A 62-year-old patient with a large HCC and main PVT who underwent treatment intensification (boosted). The patient showed a major response with revascularization of the portal vein. a–c Initial imaging: CT slices show a heterogeneous HCC of 9.6 cm (a) with main PVT (b); MAA SPECT/CT image shows high uptake in the main PVT (c) The patient underwent treatment intensification with 1.16 GBq of ⁹⁰Y-loaded glass microspheres (ILD 211 Gy, TD 285 Gy, HILD 65 Gy). Using the standard approach only 0.56 GBq of ⁹⁰Y-loaded glass microspheres would have been used to achieve an ILD of 120 Gy, then the TD would have been only 162 Gy and the HILD 37 Gy. d, e CT slices 3 months after injection show EASL partial response of the tumour (d) and main portal vein revascularization (e). The patient subsequently received a left hepatectomy (with complete tumour resection). PFS was 15 months with lung recurrence only. The patient was still alive at the most recent follow-up visit (18 months)

Fig. 3

Kaplan-Meier estimates of TTP (a) and OS (b) for the whole population (n = 71) stratified by TD

Fig. 4

Kaplan-Meier estimates of TTP (a) and OS (b) in patients with PVT (n = 33) stratified by TD

Fig. 5

Kaplan-Meyer estimates of TTP (a) and OS (b) for poor (n = 5) or good PVT (n = 28) candidates

Fig. 6

Kaplan-Meier estimates of TTP (a) and OS (b) in Child-Pugh class A patients with segmental or lobar PVT (n = 18)

Fig. 7

ROC curves for the three toxicity scoring systems