The category of nonresectable GIST covers a variety of distinct entities. A primary tumor near the gastroesophageal junction, for instance, can be challenging to resect using standard laparoscopic techniques. Other examples that may be considered nonresectable include a duodenal GIST that might require a Whipple procedure, a large pelvic GIST that requires extensive pelvic surgery, a 30-cm GIST that is attached to multiple organs and involves the mesenteric vessels, or a GIST that is in both lobes of the liver and cannot be completely removed. Additionally, a GIST might be considered nonresectable as a reflection of an individual patient’s surgical risk rather than any specific surgical limitation. For example, in a patient who has had multiple heart attacks or has uncontrolled diabetes, even a simple resection might have an unacceptable risk of complications/death.

Regarding surgery for more advanced GIST, multidisciplinary input is very helpful. When we think about surgery in the metastatic disease setting (eg, cytoreductive surgery), it is important to remember that medical therapy prior to surgery should be considered for a number of reasons. First, we would like to define our optimal therapy. If, for some reason, the GIST is resistant to our first therapy, we would proceed to define an effective second-line therapy. Thus, even in patients with metastases that we believe are fully resectable, initial treatment with imatinib may be useful to determine whether the patient will develop resistance, since medical treatment will still be required after cytoreductive surgery. Additionally, it is far easier to perform surgery on tumors that have been treated with an effective therapy; they become smaller, less vascular, and less friable, enabling surgery to proceed more easily and with less morbidity.

Whether we are dealing with a primary GIST that is nonresectable or an initial presentation of a metastatic GIST, we would usually start by identifying the driver mutation. In the case of a nonresectable KIT exon 9–mutated GIST, we would usually treat the patient with imatinib 800 mg per day, or the highest dose that the patient is able to tolerate while maintaining a reasonable quality of life. We would then typically perform scans with contrast-enhanced computed tomography, typically once every 2 to 3 months. At the point of maximal response, with the greatest reduction in tumor size, we would present the case again to our multidisciplinary GIST board and discuss whether any surgical options for the patient have emerged. Sometimes a GIST that is initially considered nonresectable has encroached on or has pushed against an organ (eg, the spleen or the liver) but has not actually invaded that organ. Targeted therapy with imatinib, for instance, can facilitate surgical resection in some of these cases. In the case of a PDGFRA exon 18 mutation, avapritinib was recently approved by the US Food and Drug Administration for that specific indication. If a different driver mutation were present (eg, NTRK translocation, RAS mutation), we would apply the same principles. 

Succinate dehydrogenase–deficient GIST, which arise nearly exclusively in the stomach, are generally resistant to imatinib; additionally, sunitinib and regorafenib are not very effective against these mutations. We typically do not see a robust decrease in tumor size when these patients are treated, and gastrectomy has not been recommended. So, a segmental resection or a wedge resection would seem to be more favorable than a complete gastrectomy, even in the setting of microscopically positive margins. As GIST is a rare disease, clinical trials should always be considered at each line of therapy. This is true of all GIST, especially succinate dehydrogenase–deficient GIST.

The main reason for neoadjuvant therapy, as Dr Trent and Dr Heinrich mentioned, is cytoreduction, which can render nonresectable disease resectable. Before treatment with imatinib, GIST is often soft, friable, and highly vascular, and it is easy for these tumors to rupture. However, in many cases, treatment with imatinib can improve the ability to manipulate these tumors. While neoadjuvant therapy is important to consider, there is no level 1 evidence for this practice at this time, but, as the evidence base grows, we are hopeful that this may become the standard of care. This also dovetails with Dr Trent’s point regarding the rarity of the disease and the importance of clinical trials.

Turning to more advanced GIST, most patients with GIST who are on second- or third-line therapies, or beyond, have metastatic disease, which may involve several areas of the abdomen and pelvis. These are often patients who have been on imatinib, and the disease came back. In this scenario, you prescribe sunitinib or regorafenib. Surgery in the context of recurrent and/or multifocal metastatic disease certainly is something that should be discussed among a multidisciplinary tumor board consisting of a surgeon, radiation oncologists, radiologists, and pathologists, among other specialists. I agree that, in certain cases, surgery does make sense, but it really needs to be done judiciously. A particular scenario that falls in this category is that of a patient who is perhaps on second- or third-line therapy and perhaps has 5 spots in the abdomen, and everything is controlled well with regorafenib, for instance. And then, after 1 year of treatment, 1 out of the 5 spots grows 1 cm, while everything else is stable. We would consider cases like this to be a clonal outgrowth, and it would be reasonable to approach that with surgery, radiation therapy, or some other option for local control.