This was an eye-opening study by Murphy et al who looked at patterns of response of hepatoblastoma (both volume reduction and regression away from major vascular structures) treated with neo-adjuvant chemotherapy.
It was eye opening for two reasons.
The first reason is that one would assume that substantial shrinkage of tumor would translate into substantial regression away from major vascular structures (and subsequently facilitate resection). It turns out that although these tumors respond well to neo-adjuvant chemotherapy, they shrink towards major vascular structures (as if anchored to their main blood supply), and not centripetally. In fact, the authors noted that the tumors, at most, regressed an average of 1 mm from major vascular structures.
The second, and more important eye opener, was the remark made by Dr Meyers in the discussion section of the paper. The authors were addressing the recommendation by the COG protocol (AHEP0731) that patient with margins less than 1 cm from major vascular structures not undergo attempts at primary resection but receive neo-adjuvant chemotherapy. The authors conculded that chemotherapy brought only a few patients closer to respectability (because it brought only a few patients to this 1 cm level)
The misinterpretation of COG recommendations lies in the following: The 1 cm recommended margin away from major vascular structures is intended to encourage neoadjuvant chemotherapy and avoid positive margins with attempts at primary resection. So really, the < 1 cm margin is more an indication for chemotherapy then a contraindication for surgery.
Subsequently, once a patient has received neo-adjuvant chemotherapy, the entire < 1 cm business is out the window, and resection should be attempted if deemed possible and safe regardless of radiologic vascular margins.