Reply to Riquet et al.

We thank Riquet et al. for their kind and meaningful comments on our manuscript [1, 2]. In an attempt to explain the different prognostic significance of micro-N2 between their study [3] and ours [2], they have questioned the impact of neoadjuvant therapy (NAT) on micro-N2. In our cohort, the survival of micro-N2 was similar, regardless NAT was given or not [median overall survival (OS): 41 months (95% CI 32.7–49.3) vs 43 months (95% CI 35.26–50.74), P = 0.08, respectively]. This observation is not surprising, considering the marginal impact of NAT onto survival; addressing a category of improved prognosis, we hypothesize that the relatively weak effect of NAT may not reveal itself on a limited sample size. Opposite results might also be explained by dissimilar sampling sizes (53 patients in the publication by Riquet et al. vs 309 in ours) leading to a lack of power. In addition, we have to be cautious about differences in patient selection and a potential Will Rogers effect. Our cohorts are not comparable; Riquet et al. divided their population into three categories: non-bulky N2 (below 2 cm, G2 group), bulky N2 (above 2 cm, G3 group) and micrometastases (G1 group). The authors did not detail criteria defining the latter group. Hence, we may speculate that G2 group, which harbours the best prognosis, may have included patients with N2 ranging from 0.2 to 2 mm, improving OS of this group. On the other hand, our non-micro-N2 group has included all the patients with metastasis >2 mm (including consequently G2 and G3 groups), which may change prognosis by an effect of different pooling.

Riquet et al. [3] argued that the adverse prognosis of micro-N2 is an argument for a more aggressive disease. However, micro-N2 is a necessary initial step leading to macroscopic N2, and its diagnosis cannot herald natural history of the disease. However, cancer cells harbouring EGFR mutations are higher responders to radiotherapy and chemotherapy [4]. We have shown in a subsequent study that micro-N2 is correlated with EGFR mutations (manuscript in preparation), which exhibits better prognosis in Caucasian surgical cohorts [4].

In conclusion, prospective studies are necessary to evaluate the prognostic value of micro-N2.

REFERENCES