What kind of tumour rupture requires adjuvant therapy?

We have read with great interest an original article by Heikki Joensuu et al., which retrospectively investigated the impact of adjuvant imatinib therapy for gastrointestinal stromal tumour (GIST) ruptures based on data from a randomised controlled trial (SSGXVIII/AIO), in which subjects with high-risk GIST were randomly assigned to receive adjuvant imatinib therapy for either 1 year or 3 years postoperatively [1]. The authors concluded that despite recurrence-free survival (RFS) and overall survival (OS) being poorer in the trial subjects with a ruptured GIST than in the other, high-risk subjects, a commendable 10-year OS rate was achieved in the 3-year adjuvant imatinib cohort when the GIST harboured an imatinib-sensitive KIT exon 11 deletion/indel mutation [1].

Our concern revolves around the criteria for defining a GIST rupture. It is now generally accepted that the prognosis of patients with a GIST rupture is much bleaker than that of other types of high-risk GIST patients, yet the definition itself is nebulous owing to its variation across studies. The Oslo criteria published by Hølmebakk et al. in 2016, which categorised GIST ruptures into major and minor types, have been adopted by a growing number of researchers [2]. Nishida et al. later introduced a modified definition of GIST rupture in 2019 that marked a refinement of the Oslo criteria by excluding four, minor rupture patterns, such as those produced by a core- or fine-needle biopsy without complications [3]. It is noteworthy that when selecting their cases of GIST rupture, Heikki Joensuu et al. chose to ignore Nishida et al.’s new definition and instead used their own modification of the Oslo criteria. Specifically, the authors did not include ‘microscopic direct invasion into adjacent organs’ in their criteria, adopting instead three types of minor defect besides intraluminal tumour perforation as instances of a minor rupture, thus exemplifying the lack of a global consensus on the definition of GIST rupture. Moreover, their distinction between a major and minor rupture, particularly in terms of visible leakage of the tumour content into the peritoneal cavity, was not based on objective criteria. A significant number of cases in their study (12 out of 69) involved a rupture of unclassified degree; this group may have been classified as a major or minor rupture due to the 'subjective judgement' of the surgeon or the potential recall bias associated with the retrospective assessment of past surgical records, which may have altered their study results. Hølmebakk et al. have shown that the prognosis of GIST with a minor rupture did not differ significantly from that of GIST without a rupture [2, 4, 5]. Conversely, while Heikki Joensuu et al. included only nine patients with a minor rupture, the prognosis of these patients was as poor as that of patients with a major rupture. Needless to say, these facts inject considerable uncertainty into their findings. A GIST rupture may be relatively rare, but how it is defined can have a significant impact on outcomes.

Another concern is the timing of the ruptures. Their study included both preoperative and intraoperative ruptures but failed to compare their prognosis. As we previously reported, intraoperative GIST ruptures do not necessarily have a poor prognosis [6]. Peritoneal metastasis of a sarcoma is relatively rare, and the underlying, molecular mechanism is still unknown. However, as with malignant, epithelial tumours, peritoneal dissemination of GIST may involve a complex process that includes adhesion, invasion, and proliferation. Therefore, it is counterintuitive to treat preoperative (spontaneous) ruptures lasting as long as several weeks in the same manner as intraoperative (iatrogenic) ruptures lasting a much shorter time. Furthermore, a peritoneal lavage may affect the outcome, although the decision to perform the procedure is left to the discretion of the surgeon, and surgical records usually do not mention lavage. Numerous studies have consistently shown that tumour ruptures correlate with an increased risk of recurrence, affirming the necessity for adjuvant imatinib therapy exceeding 3 years in patients with a GIST rupture [3]. However, further investigation is required to determine if minor as well as intraoperative ruptures should be treated as GIST ruptures requiring 3 years or more of adjuvant imatinib therapy; long-term adjuvant imatinib therapy is physically, mentally, and financially burdensome, and should be avoided in patients who do not need it.

In conclusion, we emphasize that the definition of a GIST rupture requiring prolonged adjuvant imatinib therapy is still uncertain, and discretion should be exercised when applying the results of Heikki Joensuu et al.’s study to clinical practice.

Author contributions

Both GC and HC wrote, revised and approved the final version of the manuscript.

Competing interests

The authors declare no competing interests.