Both Mullen et al. and Peng et al. report on retrospective series of patients affected by desmoid-type fibromatosis, collected over decades, to investigate prognostic factors for outcome after surgical resection. Although not new, their conclusions add to the ongoing debate on this issue and contribute to the understanding of this disease, the approach to which is changing worldwide and should be properly codified. Interestingly their conclusions diverge. On multivariable analysis surgical margins are prognostic of local recurrence in the series by Mullen et al., but not in the other. On the other hand, on multivariable analysis tumor site of origin and age are prognostic of local recurrence in the series by Peng et al., but not in the other. Thus, these 2 manuscripts reflect the contradictory results found in the literature. Indeed optimal management of desmoid-type fibromatosis has not yet been defined, but surgery, if feasible, is still largely felt to be the mainstay of treatment, albeit followed by a significant local failure rate. This being true, the first observation is that the failure rates found by both authors are somewhat higher than those from other major series (Table 1). The expected actuarial failure rate after surgery should be approximately 30 % at 10 years, whereas these 2 series report 40 and 50 %, respectively. One possible explanation may be related to the relative short median follow-up of both (40 and 25 months, respectively). In fact, given the overall benign nature of this disease, it is possible that patients who have not recurred within a short period postoperatively may have left the follow-up program relatively early or are continuing follow-up at centers closer to their homes rather than at these referral centers. Therefore, they may be censored relatively early despite having long disease-free intervals that is not reflected in the recurrence-free survival curve. In other words, the recurrence-free survival curves may be biased toward those developing recurrences. From a methodological standpoint, such an indolent disease should be studied in series with long follow-up times for both those who have recurred and those who have not, for a more accurate representation of recurrence rates. Surgery is an option able to control this disease in 70 % of the cases. The challenge is to select the proper patients to offer a successful treatment. The second issue that these manuscripts raise is one that has been a long-lasting question—when a decision for surgery is made, what are appropriate margins of resection? If they are prognostically significant, then the goal of surgery should be achieving negative margins and surgical morbidity should be carefully balanced between risk of recurrence and morbidity. If not, a more relaxed approach to functional structures and function preservation may be pursued. The propensity of aggressive fibromatosis to locally recur should be related to its infiltrative nature. A positive margin clearly means that active tumor has been left behind. Why has this issue of appropriate margins remained such a persistent area of debate? One possible explanation is that the reported results may have been biased by some factors, such as the inclusion in the same series of both intra-abdominal and extra-abdominal tumors, sporadic, and FAP-associated ones, and both primary lesions and recurrences. In addition, patients may have received disparate treatments. Indeed this was also the case Society of Surgical Oncology 2012