Results confounded by a disregard for basic dose-response radiobiology.

REPLY: We would like to thank Drs. Lam and Smits for their concerns and comments regarding the methodology in our study (1). The aim of our study was to answer the frequently occurring clinical question of whether a patient with low or no 99mTcmacroaggregated albumin (99mTc-MAA) uptake in metastatic lesions should undergo 90Y-radioembolization. The observation in our patient cohort with colorectal liver metastasis was that therapy response after 90Y-radioembolization was independent of the degree of intratumoral 99mTc-MAA uptake. Consequently, our recommendation to the reader was that “therapy should not be withheld from patients with colorectal liver metastases lacking intratumoral 99mTc-MAA accumulation” (1). Our results are based on the current body-surface-area model available, taking all the insufficiencies and drawbacks of the surrogate 99mTc-MAA into account (2). The establishment of dose–response relationships was beyond the scope of our study. Although qualitative Bremsstrahlung or 90Y-PET imaging may be feasible in clinical routine, one has to admit that a quantitative assessment of dose estimations in normal liver parenchyma in regard to liver-related adverse events and in multiple tumor lesions in both liver lobes is far more difficult (3,4). However, we agree with Drs. Lam and Smits that it would be essential to establish individualized treatment planning on the basis of optimized scout-dose imaging. Besides the technical aspects, such as catheter tip position or injection flow, it is desirable to have an agent that is identical to or that better models the treatment device. The recently introduced 166Ho-microspheres by Smits et al. (5) may be used for pretherapeutic assessment and treatment evaluation, making them a promising candidate for future application. Nevertheless, we consider flow alterations during the radioembolization process due to the embolization effect to be a significant contributor to variable microsphere distribution in the tumor and liver that cannot be estimated or overcome by any proposed approach. An optimization of dose estimation and individual treatment planning is even more important for further evaluation of the clinical and biologic aspects of the dose–response relationship for different tumor entities, pretreatment with chemotherapeutics, or a combined treatment and sequential lobar treatment versus whole liver treatment (6). An individualized dosimetry concept should improve the efficacy of 90Y-radioembolization while potentially reducing cases of overtreatment and unnecessary toxicity. To define the method and role of individualized pretreatment planning, a prospective multicenter trial would be needed. Again, we thank Drs. Lam and Smits for their comments and discussion.