Perioperative Treatment of Gastrointestinal Stromal Tumors
Perioperative Treatment of Gastrointestinal Stromal Tumors
ABSTRACT: This review describes the current multidisciplinary management of gastrointestinal stromal tumor (GIST), which is the most common sarcoma of the gastrointestinal tract. Before 2001, surgery was the only effective therapy for GIST. The discovery of the central role of KIT proto-oncogene mutations in the pathogenesis of this tumor, and the development of specific inhibitors of KIT tyrosine kinase (TK) function, has changed the paradigm of treatment for GISTs. Imatinib and sunitinib are TK inhibitors with activity against GISTs. Their major established role in GIST is in the treatment of advanced disease. A growing body of literature and clinical experience support the potential perioperative use of these drugs. The adjuvant use of imatinib is based on retrospective series and limited prospective studies demonstrating that imatinib reduces the risk of recurrence. Ongoing studies are further defining the length of adjuvant therapy, as well as identifying the patients that could achieve the best results. Neoadjuvant treatment often decreases the tumor size, allowing a less morbid surgery, appears to be safe and beneficial for some patients, and therefore deserves further study.
Gastrointestinal stromal tumors (GISTs) originate from the interstitial cells of Cajal or a precursor and are the most common mesenchymal neoplasms of the gastrointestinal (GI) tract. Although GISTs often present as localized masses, they are typified by a high risk of metastatic relapse, most commonly in the liver and peritoneum. The pathologic characterization of GIST as a unique form of GI sarcomas was first described in 1983 by Mazur and Clark. Subsequently Nishida et al, Hirota et al, and later Heinrich demonstrated that GISTs often possess activating mutations in the KIT or platelet-derived growth factor receptor (PDGFR) proto-oncogenes. The identification of these mutations and their relationships with the genesis and progression of GISTs allowed for the development of effective, targeted therapeutic drugs, strongly improving the prognosis of this disease.[6,7]
Knowledge about the epidemiology of GIST has advanced markedly over the past few years. However, the true frequency of these tumors is unknown. Epidemiologic data are difficult to interpret because the early definition of GIST was derived from criteria published in 1990, and before this, GISTs were often histologically misclassified as other mesenchymal neoplasms derived from soft tissues, especially leiomyosarcomas.
In general, GISTs tend to have low growth rates, with a subsequent long time to the development of symptoms. It is no surprise, therefore, that small GISTs may remain undetected for protracted periods of time and may never progress to symptomatic or metastatic disease.
The most reliable data are derived from recent population-based studies. Investigators in Iceland examined all patients diagnosed in that country with a gastrointestinal mesenchymal tumor from 1990 to 2003. These patients were evaluated with an immunohistochemical panel including staining for c-kit, confirming an incidence rate of 11 per million. Similarly, Swedish scientists investigated all potential cases of GIST diagnosed between 1983 and 2000, reporting an incidence rate of 14.5 per million, and a prevalence rate of 129 per million.
Japanese investigators studied 100 patients whose whole stomachs were resected due to gastric cancer. The organs were sectioned at 5-mm intervals and examined by light microscopy and immunochemistry for KIT (CD117). They found 50 microscopic GISTs in a sample of 35 individuals, and 90% were located in the upper stomach. The authors suggested that, considering the annual incidence of clinical GIST, only a few microscopic GISTs grow into a clinically meaningful size with malignant potential.
A study based on the US Surveillance, Epidemiology and End Results (SEER) registry data from 1992 to 2000 found the age-adjusted yearly incidence of GIST to be 6.8 per million, although that number has been questioned as being too low. The incidence of GISTs has substantially changed over time, possibly due to the constant evolution of the diagnostic criteria and the improvement in diagnostic methods, which allows for earlier and more complete detection.
Estimation of Recurrence Risk
The estimation of recurrence risk following complete resection of GIST is of paramount importance when selecting patients who could possibly benefit from adjuvant therapy. In the past decade, several different criteria have been proposed to classify malignant potential. Although the initial diagnosis of GIST may rely heavily on molecular characteristics, the mitotic rate and tumor size, with or without location of the primary tumor, have gained the greatest acceptance as being predictive of outcome. Certainly the terms benign and malignant should not be applied to GISTs, since they all have the potential to recur and/or metastasize.
In a landmark study, DeMatteo et al reported on 200 GIST patients referred to Memorial Sloan-Kettering Cancer Center, 80 of whom had localized disease and underwent a complete resection. A total of 65 (81%) of these patients had negative microscopic margins. All patients were followed prospectively. In patients who underwent complete resection of gross disease, the 5-year survival rate was 54%. Data regarding the pattern of recurrence were available for 27 patients. Local recurrence alone occurred in 9 (33%), metastasis alone in 13 (48%), and both in 5 (19%) patients. Recurrence was predominantly intra-abdominal. On multivariate analysis, only tumor size greater than 10 cm was a significant risk factor for recurrence, with a relative risk of 2.5 (confidence interval [CI] = 1.2–5.5).
Recently, the same authors published an extended review with similar results regarding tumor size, but this time exploring the mitotic index, which is considered another classic risk factor. After a median follow-up of 4.7 years, recurrence-free survival was 83%, 75%, and 63% at 1, 2, and 5 years, respectively. On multivariate analysis, recurrence was predicted by > 5 mitoses/50 high-power fields (HPF), tumor size > 10 cm, and tumor location (with the worst outcomes in patients with small bowel GIST). A high mitotic rate conferred the most important risk factor, with a hazard rate of 14.6 (95% CI = 6.5–32.4).
The Armed Forces Institute of Pathology studied 1,765 cases of gastric GIST and demonstrated that the outcome was strongly related to tumor size and mitotic activity. Less than 3% of tumors smaller than 10 cm and with fewer than 5 mitoses/50 HPFs metastasized, whereas 86% of tumors > 10 cm and > 5 mitoses/50 HPFs metastasized. However, tumors > 10 cm with mitotic activity < 5/50 HPFs and those < 5 cm with mitoses > 5/50 HPFs had a relatively low metastatic rate (11% and 15%, respectively). Tumor location in the fundus or in the gastroesophageal junction, coagulative necrosis, ulceration, and mucosal invasion were significant unfavorable factors (P < .001), whereas tumor location in the antrum was favorable (P < .001).
The same group also reported on 906 patients with small bowel GIST. In this review, tumor-related mortality was 39%—twice what was observed in the gastric GIST review. A similarly strong correlation was seen between outcome and both tumor size and mitotic activity (Table 1). Fewer than 3% of tumors < 5 cm and ≤ 5 mitoses/50 HPF metastasized, whereas distant spread was observed in 86% of tumors > 10 cm and > 5 mitoses/50 HPF. However, in stark contrast to the corresponding intermediate-risk gastric tumors, small bowel tumors > 10 cm with mitotic activity ≤ 5/50 HPF and those ≤ 5 cm with mitoses > 5/50 HPF had a high metastatic rate (> 50%). Even tumors > 5 cm ≤ 10 cm with a low mitotic rate had a 24% metastatic rate.
In general, tumors arising from the small bowel, colon, rectum, or mesentery are generally associated with less favorable outcomes than those arising from the stomach.[15,16] Thus, based on the more recent review, the most commonly used criteria to assess the risk of recurrence include the primary tumor diameter, the mitotic count, and the location of the resected tumor. Additionally, other factors may be prognostic. Patients whose tumors have ruptured into the abdominal cavity have a very high risk of tumor recurrence. Recently, Huang et al reviewed the data from 289 patients with localized GIST and proposed a modification in the consensus risk stratification. Very low-risk and low-risk GISTs were combined into a single new risk level I group, and intermediate-risk GISTs were designated as risk level II. GISTs > 5 cm with > 10 mitoses/50 HPF were designated risk level IV. All other high-risk GISTs were defined as risk level III. The cumulative 5-year rate of disease-specific survival for patients with GISTs classified as risk levels I to IV were 100%, 96%, 67%, and 25% at 5 years, respectively. Despite the relatively small study population, the differences in outcome were striking between patients in risk level II and III (P < .0001) and between those in risk level III and IV (P = .0002). These impressive results suggest that this new prognostic criteria could be superior to the commonly used National Institutes of Health consensus criteria, especially to stratify high-risk patients (Table 2).