Yunfan Sun, Xinrong Yang, Jian Zhou

Liver Cancer Institute, Zhongshan Hospital, Fudan University

Objective:Our previous study indicated that apreoperative CTC≥2 was an independent prognosticator for early recurrence inhepatocellular carcinoma (HCC). Recurrent foci developed in these patients weremostly observed near the surgical margin of remnant liver. However, there is nodata on what constitutes an adequate surgical margin in those patients.Inadequate surgical margins may result in high local recurrence due tomicrovascular invasion (mVI) by tumor cells, and excessively large resectionsmay fail to preserve more functional liver parenchyma. Herein, we investigatedthe correlation between preoperative CTC load and features of mVI and theoptimum surgical margin threshold for patients with CTC≥2 or CTC. Method: Weprospectively enrolled 26 HCC patients to evaluate the relationship between CTCcounts and features of mVI and another cohort of 179 HCC patients to examinethe optimal surgical margin among patients with preoperative CTC ≥2 or CTC. Result:In 26 patients, 20 patients had mVI detected. The proportions of patientswith farthest mVI distance of 0.5 cm, 1 cm, 2 cm, 3 cm and>3 cm were 19.2%,19.2%,15.4%, 11.5% and 11.5% respectively. Of all microvessels invaded, 45.6%distributed ≤12.5px from tumor edge, 18.4% within>0.5 to ≤1 cm, 20.41% within>1to ≤2 cm, 12.2% within>2 to ≤3 cm and 3.4%>3 cm. The number of CTC waspositively correlated with the farthest distance of mVI from the tumor edge (r=0.697,P<0.001) and the number of microvessels invaded (r=0.656, P<0.001).In 12 patients who had ≥2 CTC detected, the mean distance of mVI from tumor wassignificantly farther than those with CTC (2.29±0.41 vs. 0.64±0.19, P<0.001),and the number of microvessels invaded was significantly higher than those withCTC (9.50±1.04 vs. 2.36±2.23, P=0.006). In 179 patients, 73 had CTC of≥2 preoperatively. Among them, 55.6% had resection margin width of ≤0.5 cm,18.9% within>0.5 to ≤1 cm, 17.8% within>1 to ≤2 cm and 7.7% within>2to ≤3 cm. Compared with a margin of>1 to ≤2 cm, patients with margin widthof ≤0.5 cm (81.6% vs. 38.5%, P=0.003) and of>0.5 to ≤1 cm (81.6% vs.38.5%, P=0.048) were associated with a significantly higher risk ofpostoperative recurrence. However, when a>1 to ≤2 cm margin was compared witha margin of>2 to ≤3 cm (38.5% vs. 33.3%, P=0.924), no significantdifference in the risk of recurrence was observed. Therefore, a 1-cm margin wasdesignated as optimal threshold surgical margin for patients with CTC≥2 atpreoperation. Patients with a margin of>1 cm had reduced rates of recurrence(35.3% vs. 78.5%) and shorter time to recurrence (median 3.31 months vs. notreached, P=0.002) compared with those with a ≤1 cm margin. In 106patients with CTC 1 cm and ≤1 cm (31.2% vs. 34.5%, P=0.709). Moreover,when a ≤0.5 cm margin was compared with a margin of>0.5 to ≤1 cm in patientswith (31.6% vs. 20%, P=0.338), no significant difference in recurrencerisk was observed. Conclusion: The load of CTC before surgery positivelycorrelated with the number and distance of mVI. In patient with ≥2 CTCdetected, a threshold of 1 cm is the optimal width of surgical margin, whichmay effectively get rid of two-thirds of peritumoral mVI. A 1-cm surgicalmargin not only has a similar impact on reducing recurrence risk as a largermargin do, but also facilitates the preservation of functional liverparenchyma.

Key Words: Ciruclating tumor cells  Hepatocellular carcinoma