Tumoricidal dosing approach with parenchymal sparing using voxel-based dosimetry in 90 Y glass microspheres treatment of hepatocellular carcinoma.

OBJECTIVE

We aimed to evaluate the effect of tumor absorbed doses (TAD) on treatment response in patients with hepatocellular cancer (HCC) treated with 90 Y glass microspheres. We aimed to define a cutoff value for complete response (CR).

METHODS

The voxel-based dosimetry for the treatment of 66 HCC lesions in 56 patients was analyzed retrospectively. Nineteen patients had BCLC A, 23 patients had BCLC B, and 14 patients had BCLC C disease. Treatments were grouped as selective (radiation segmentectomy and super-selective segmentectomy, n:49) and nonselective (palliative treatments for tumors occupying >2 segments, n:17). Treatment response was evaluated by mRECIST criteria, defined as CR, partial response (PR), stable lesion (SL), and progressive lesion (PL). TAD associated with CR was analyzed.

RESULTS

TAD was 525 ± 222 Gy in our cohort. Fifteen lesions had CR, 28 had PR, eight remained stable, and 15 lesions progressed. CR, PR, SL, and PL rates for selective vs. nonselective treatments were 31, 42, 12, and 14% vs. 0, 41, 11, and 47% for nonselective treatments, respectively ( P :0.01). TAD was significantly associated with treatment response. Receiver operating characteristic analysis showed TAD > 475 Gy predicted CR with 100% sensitivity and 68% specificity (area under the curve = 0.83, P  < 0.001). Overall survival declined as treatment response deteriorated. None of the patients had radiation-induced liver dysfunction on follow-up (6-21 months).

CONCLUSION

Higher TAD is crucial for CR. Segmentectomy with TAD > 475 Gy is associated with favorable response and better survival in HCC patients. Even for palliative treatments, as high as reasonably tolerated doses must be applied to achieve a favorable response.