Risk Criteria and Prognostic Factors for Predicting Recurrences After Resection of Primary Gastrointestinal Stromal Tumor

Transcription

1 Annals of Surgical Oncology 14(7): DOI: /s Gastrointestinal Oncology Risk Criteria and Prognostic Factors for Predicting Recurrences After Resection of Primary Gastrointestinal Stromal Tumor Piotr Rutkowski, MD, PhD, 1 Zbigniew I. Nowecki, MD, PhD, 1 Wanda Michej, MD, 2 Maria Dębiec-Rychter, MD, PhD, 3 Agnieszka Woz niak, PhD, 4 Janusz Limon, PhD, 4 Janusz Siedlecki, PhD, 5 Urszula Grzesiakowska, MD, PhD, 6 Michał Kąkol, MD, 7 Czesław Osuch, MD, PhD, 8 Marcin Polkowski, MD, PhD, 9 Stanisław Głuszek, MD, PhD, 10 Zbigniew _Zurawski, MD, 1 and Włodzimierz Ruka, MD, PhD 1 1 Department of Soft Tissue/Bone Sarcoma and Melanoma, M. Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena 5, , Warsaw, Poland 2 Department of Pathology, M. Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena 5, , Warsaw, Poland 3 Center for Human Genetics, University of Leuven, O&N Gasthuisberg Herestraat 49, 3000, Leuven, Belgium 4 Department of Biology and Genetics, Medical University of Gdansk, Debinki 1, , Gdansk, Poland 5 Department of Molecular Biology, M. Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena 5, , Warsaw, Poland 6 Department of Radiology, M. Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena 5, , Warsaw, Poland 7 Regional Oncological Center, Marii C. Sklodowskiej 2, , Gdansk, Poland 8 Department of General Surgery, Jagiellonian University, Kopernika 40, , Cracow, Poland 9 Department of Gastroenterology, M. Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena 5, , Warsaw, Poland 10 City Hospital, Grunwaldzka 45, , Kielce, Poland Background: The introduction of adjuvant imatinib in gastrointestinal stromal tumors (GISTs) raised debate over the accuracy of National Institutes of Health risk criteria and the significance of other prognostic factors in GIST. Methods: Tumor aggressiveness and other clinicopathological factors influencing diseasefree survival (DFS) were assessed in 335 patients with primary resectable CD117-immunopositive GISTs (median follow-up, 31 months after primary tumor resection) from a prospectively collected tumor registry. Results: Overall median DFS was 37 months, and estimated 5-year DFS was 37.8 %. In univariate analysis, high or intermediate risk group (P < ), mitotic index >5/50 highpower field (P <.00001), primary tumor size >5 cm (P <.00001), nongastric primary location (P =.0001), male sex (P =.01), R1 resection/tumor rupture (P =.0003), and epithelioid cell or mixed cell pathological subtype (P =.05) negatively affected DFS. In multivariate analysis, statistically significant factors negatively influencing DFS for model 1 were mitotic index >5/50 high-power field (P =.004), primary tumor size >5 cm Received December 15, 2006; accepted January 17, 2007; published online: May 2, Address correspondence and reprint requests to: Piotr Rutkowski, MD, PhD; rutkowskip@coi.waw.pl. Published by Springer Science+Business Media, LLC Ó 2007 The Society of Surgical Oncology, Inc. 2018

2 PREDICTING RECURRENCE IN PRIMARY GIST 2019 (P =.001), male sex (P =.003), R1 resection/tumor rupture (P =.04), and nongastric primary tumor location (P =.02), and for model 2 were high/intermediate risk primary tumor (P <.0001 and P =.008, respectively), male sex (P =.007), resection R1/tumor rupture (P =.01), and nongastric primary tumor location (P =.02). Five-year DFS for high, intermediate, and low/very low risk group was 20%, 54%, and 96%, respectively. Conclusions: The risk criteria for assessing the natural course of primary GISTs were validated, but additional independent prognostic factors primary tumor location and sex were also identified. Key Words: Gastrointestinal stromal tumor Surgery Prognosis. Gastrointestinal stromal tumors (GISTs) comprise a recently defined entity of the most common mesenchymal neoplasms of the abdominal cavity. GISTs are believed to arise from precursors shared with the interstitial cells of Cajal, the gut pacemaker cells. 1 3 Activating somatic KIT mutations occur in approximately 80% of GISTs, resulting in overexpression of the KIT receptor protein and common CD117 immunopositivity. 3 5 However, considerable controversy exists regarding GISTs prognosis and outcomes after primary surgery. GISTs are morphologically and clinically heterogeneous tumors, for which the biological behavior is difficult to predict, ranging from clinically benign to malignant. 6 A consensus conference held at the National Institutes of Health (NIH) in provided both an evidence-based definition and a practical scheme for the assessment of the risk in the clinical course of this disease (Table 1). The risk categorization is based on evaluation of the size and mitotic rate of the tumors as the most reliable prognostic factors, and its use is strongly advocated. However, the design of trials assessing adjuvant treatment in GIST patients with imatinib has raised debate about the accuracy of the NIH consensus criteria for the risk of disease recurrence after surgical treatment of primary tumors. 7,8 The aim of our study was to analyze the value of the NIH criteria and other prognostic factors, as well as clinicopathologic features, in the relationship to disease recurrences in a large series of primary GIST patients enrolled onto our tumor registry. PATIENTS AND METHODS We prospectively analyzed data from 335 patients with primary resectable GIST and no evidence of metastatic disease enrolled onto Clinical GIST Registry from In all patients, macroscopically radical resection of the tumor was performed. Clinicopathologic data was supplemented by a review of all available medical and histopathological records TABLE 1. Risk of aggressive behavior in gastrointestinal stromal tumor according to the National Institutes of Health criteria Risk category Tumor size (cm) a MI (/50 HPF) b Very low risk <2 5 Low risk Intermediate risk > High risk >5 >5 >10 Any MI Any size >10 MI, mitotic index; HPF, high-power field. a Size represents the single largest dimension. b Ideally, MI should be standardized according to surface area examined (based on size of HPF). from the referring hospitals. The diagnosis of GIST was confirmed by histopathological review of all cases, and by immunohistochemical staining for CD117 (antibodies purchased from Dako, Carpintiera, CA), which was performed in the Department of Pathology, Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology in Warsaw, Poland. Imatinib was not used as an adjuvant treatment in any of these patients. Patients did not undergo any further selection. Follow-up information was obtained during regular outpatient visits or by phone with the patient and/or the referring physician. Postoperative follow-up consisted of physical examination and standard imaging (computed tomography of the abdominal cavity and the pelvis, and chest x-ray). Thorough examinations were routinely performed every 3 months during the first 2 years, every 4 months during the third year, and every 6 months in the years afterward. Mutational analysis of KIT and platelet derived growth factor receptor alpha (PDGFR-a) was performed in randomly selected 90 cases on the basis of DNA isolated from paraffin-embedded or fresh frozen tumor tissue. KIT exons 9, 11, 13, and 17, and PDGFR-a exons 12 and 18 were amplified by polymerase chain reaction and prescreened with denaturing high-performance liquid chromatography (WAVE System, Transgenomics). Samples with abnormal

3 2020 P. RUTKOWSKI ET AL. elution profiles were bidirectionally sequenced, as previously described. 9 Molecular analysis study had been approved by the local bioethics committee according to good clinical practice guidelines. During follow-up, we analyzed the incidence and pattern of disease recurrences. Disease-free survival (DFS) was calculated from the date of primary tumor resection to the date of recurrence or the last followup date. All deaths from other causes were recorded as censored. DFS was assessed with respect to the following variables: demographic data (age at diagnosis 45 years vs. >45 years, and sex), tumor size ( 5 cm vs. 5 to 10 cm vs. >10 cm), mitotic rate (0 to 5 vs vs. >10 per 50 high-power fields [HPF]), NIH consensus risk criteria 4 (combining tumor size and mitotic rate: very low/low risk vs. intermediate risk vs. high risk), histological subtype (spindle cell vs. epithelial cell or mixed cell), primary tumor site (gastric vs. small bowel vs. large bowel vs. others), type of surgical resection (R0 microscopically radical resection vs. R1 microscopically nonradical, but macroscopically radical resection or tumor intraoperative rupture), molecular findings (KIT exon 11 mutants vs. others). We did not perform an analysis for overall survival because most patients had been treated with imatinib after disease recurrence. All statistical analyses were performed by SAS software (SAS Institute, Cary, NC), version 8.2, and Statistica software (Statsoft, Tulsa, OK). Contingency tables were analyzed by the v 2 test for nonnormal distribution of parameters. Groups were compared for age differences by the Mann-Whitney U-test. For the survival analysis, Kaplan-Meier estimates were used with generalized Wilcoxon and the log rank tests for bivariate comparisons. In multivariate analysis of the factors associated with DFS after radical surgery of primary GISTs, we used the Cox proportional hazard models, applying the stepwise model building procedure that included all covariates significant at the 20% level in bivariate analysis. Two-way interactions were then considered in the model. The scale of numerical covariates was examined by the martingale residuals based methods and fractional polynomial method, with two terms and powers in the set {)2, )1, ).5,.5, 1, 2, 3} and the log of the variables. To detect the lack of fit of the considered models, Gronesby and BorganÕs test was used. 10 Differences were considered statistically significant if P values were <.05. The goals of our analysis were to validate the NIH consensus risk group criteria and to identify other potential prognostic factors. To accomplish this, we first fitted a model containing the NIH consensus categories for mitotic index (MI) and primary tumor size (model 1). The next step was to include the NIH consensus risk categories (model 2). Finally, the third approach was to include the number of mitoses (/50 HPF) and the tumor size (cm) as continuous covariates (model 3). RESULTS Patient Characteristics The group analyzed consisted of 335 patients (156 male [47%] and 179 female [53%]) with a median age at diagnosis of 58 years (range, 9 89 years). One hundred four cases with primary inoperable and/or metastatic disease at presentation (104 [24%] of 439 of all GIST cases in our database; data not shown) were not included in the present study. The detailed clinicopathological data of the entire group of patients with primary GISTs are shown in Table 2. The most common symptom at presentation was a palpable intraabdominal mass (49%; 169 cases), followed by gastrointestinal bleeding or anemia (18.5%; 62 cases) and abdominal pain (18%; 60 cases). In 44 patients (13%), the primary operation was performed as an emergency procedure required by perforation of the gastrointestinal tract, bowel obstruction, or gastrointestinal bleeding. In 10 patients (3%), GISTs were resected during surgical procedures performed for other reasons. The diagnosis of GIST was established preoperatively in only 20 patients (6%) (these lesions were localized in the stomach in all but one case). Lymph node metastases were rare and occurred only in four cases (1.2%). In two patients, primary gastric GISTs were part of the Carney triad, with coexisting pulmonary multiple chondromas. Three patients with primary small bowel GISTs showed clinical symptoms of neurofibromatosis type 1. We found that 29 patients (8.7%) had a history of other malignant neoplasms (24 with metachronous and 5 synchronous neoplasms). Patterns and Time of Recurrence of the Disease Median follow-up time was 31 months (range, months). During follow-up, we detected 151 cases of disease recurrences (45.1%). Ten (6.6%) were local recurrences, 9 (6.0%) were local recurrences with liver metastases, 38 (25.2%) were liver metastases only, 1 (.6%) was a lung metastases only, 50 (33.1%) were intraperitoneal dissemination, and 43 (28.5%) were intraperitoneal disseminations and liver metastases.

4 PREDICTING RECURRENCE IN PRIMARY GIST 2021 Clinicopathological feature TABLE 2. Characteristics of 335 patients with primary resectable gastrointestinal stromal tumors Total, n (%) (n = 335) RF, n (%) (n = 184) DR, n (%) (n = 151) P value (RF vs. DR) Age (y) Median (range) 58 (9 89) 62 (22 85) 53 (9 89) Mean <.001 <45 52 (15.5%) 20 (10.9%) 32 (21.1%) > (84.5%) 164 (89.1%) 119 (88.9%).009 Sex Female 179 (53.4%) 106 (57.6%) 73 (48.3%) Male 156 (46.6%) 78 (42.4%) 78 (51.7%).09 Primary tumor site Stomach 153 (45.7%) 107 (58.2%) 46 (30.5%) Duodenum 16 (4.8%) 12 (6.5%) 4 (2.6%) Small bowel 142 (42.4%) 57 (30.9 %) 85 (56.4%) Large bowel/rectum 13 (3.9%) 4 (2.2%) 9 (5.9%) Other or intraperitoneally with 11 (3.2%) 4 (2.2%) 7 (4.6%) <.0001 unknown primary origin Primary tumor size (cm) Median (range) 8 (.7 40) 5 (.7 40) 11 (2 35) 5 99 (29.6%) 92 (50%) 7 (4.6%) > (36.4%) 51 (27.7%) 71 (47.0%) > (29.8%) 38 (20.7%) 62 (41.1%) Missing data 14 (4.2%) 3 (1.6%) 11 (7.3%) <.0001 MI (/50 HPF) Median (range) 4 (0 100) 2 (0 100) 10 (0 100) (51.1%) 136 (73.9%) 35 (23.2%) (14%) 20 (10.9%) 27 (17.9%) >10 76 (22.7%) 12 (6.5%) 64 (42.4%) <.0001 Missing data 41 (12.2%) 16 (8.7%) 25 (16.5%) Risk group Very/low 78 (23.3%) 76 (41.3%) 2 (1.3%) Intermediate 68 (20.3%) 47 (25.5%) 21 (13.9%) High 177 (52.8%) 57 (40.0%) 120 (79.5%) <.0001 Missing data 12 (3.6%) 4 (2.2%) 8 (5.3%) Pathological subtype Spindle cell 143(42.7%) 81 (44%) 42 (27.9%) Epithelial cell or mixed 64 (19.1%) 28 (15.2%) 36 (23.8%).004 Missing data 128 (38.2%) 75(40.8%) 73 (48.3%) Type of surgical resection R0 253 (75.5%) 151 (82.1%) 102 (67.5%) R1 75 (22.4%) 29 (15.8%) 46 (30.5%).001 Missing data 7 (2.1%) 4 (2.1%) 3 (2.0%) Mutational status a KIT exon 11 mutation 55 (61.1%) 24 (61.5%) 31 (60.8%) Other genetic changes 35 (38.9%) 15 (38.5%) 20 (39.2%) NS Emergency operation of primary tumor Yes 44 (13.1%) 16 (8.7%) 28 (18.5%) No 244 (72.8%) 135 (73.4%) 109 (72.2%).02 Missing data 47 (14.1%) 33 (17.9%) 14 (9.3%) RF, patients without disease recurrences during the follow-up period (recurrent-free); DR, patients with disease recurrence; R0, microscopically radical resection; R1, microscopically nonradical, but macroscopically radical resection or intraoperative tumor rupture; HPF, high-power field. a Mutation status was available in 90 cases. Overall, liver metastases were observed in 81 cases (54%). Median time to disease recurrence was 15 months (range, months). Comparisons of Groups with and without Disease Recurrence The comparison of clinicopathologic features of patients without recurrences during the follow-up period (recurrent-free [RF] group) with those who had recurrences (disease recurrent [DR] group) is summarized in Table 2. We found significantly higher median tumor size (11 cm) and median MI (10/50 HPF) in DR patients in comparison with the RF group (5 cm and 2/50 HPF, respectively) (P <.0001). The percentage of tumors with a gastric location was far lower in the DR group than in the RF group at 30.5% and 58.2%,

5 2022 P. RUTKOWSKI ET AL. FIG. 1. Disease-free survival (DFS) after primary tumor resection (N = 335). Survival distribution function and 95% confidence interval. respectively. On the contrary, the percentage of GIST located in the bowel (duodenum, small and large intestine together) was significantly higher in the DR group in comparison with the RF group (64.9% and 39.6%, respectively; P <.0001). Most patients in the DR group constituted NIH high risk cases (79.5%), followed by intermediate risk patients (13.9%). Disease recurrences were observed in only 2 (2.6%) of 78 patients from the low risk group. Mutations were detected in 71 tumors (78%), of which 55 were in KIT exon 11 (39 deletions, 6 duplications, and 10 missense mutations). We did not observe any difference with regard to the presence of detectable KIT and KIT exon 11 mutations in primary tumors between groups with versus without recurrent disease (65.8% vs. 75% and 60.5% vs. 61.5%, respectively). Factors Influencing Disease Free Survival After Primary Tumor Resection The median DFS time after resection of primary GIST was 37 months, and the estimated 5-year DFS rate was 37.8% (95% confidence interval [95% CI], ) (Fig. 1). Univariate Analysis In bivariate analysis, the following factors demonstrated negative impact on DFS: being in the high or intermediate risk group (P < ), MI >5/50 HPF (P <.00001), primary tumor size >5 cm (P <.00001), primary nongastric tumor location (P =.0001), male sex (P =.01), R1 resection /tumor rupture (P =.0003), and epithelioid cell or mixed cell pathological subtype (P =.05). The categorization of risk groups according to NIH criteria showed an excellent correlation with FIG. 2. Disease-free survival (DFS) after primary tumor resection according to National Institutes of Health risk groups. DFS (Fig. 2). DFS according to risk group indicates that the 5-year estimated DFS rate was 96% (95% CI, ) for patients with very low/low risk tumors, 54% (95% CI, ) for patients with intermediate risk tumors, and 20% (95% CI, ) for patients with high risk tumors. The statistical difference between these rates is highly significant. The 5-year DFS rate for patients with primary tumors sized 5 cm, >5 to 10 cm, and >10 cm were 76.5% (95% CI, ), 31.2% ( ), and 22.5% ( ), respectively. The 5-year DFS rate is based on the categorization of primary tumors according to MI 5/50 HPF, 6 to 10/50 HPF, and >10/50 HPF was 62.9% (95% CI, ), 23.7% ( ), and 8.0% ( ), respectively. Patients with primary gastric tumors had a significantly better 5-year DFS rate (59.7% [95% CI, ]) than patients with nongastric tumors (23.5%; 95% CI, ) (Fig. 3). Male patients had 5-year DFS rate of 33.0% (95% CI, ) compared with 42.0% for female patients (95% CI, ) (Fig. 4). Patients with microscopically nonradical (R1) resection of the primary tumor had a higher probability of disease recurrence than patients after microscopic radical resection (R0) (5-year DFS rates: 29.5% [95% CI, ] vs. 40.5% [95% CI, ], respectively). Patients with spindle cell tumors had longer DFS than patients with other histological subtypes (5-year DFS rates: 50.5% [95% CI, ] vs. 31.6% [95% CI, ], respectively). Patients with any type of KIT mutation had a similar DFS compared with the remaining patients. No statistically significant differences were observed for DFS in groups of patients with tumors harboring KIT exon 11 mutations versus others (also for subgroups of KIT exon 11 mutations: missense mutations vs. deletions/insertions).

6 PREDICTING RECURRENCE IN PRIMARY GIST 2023 FIG. 3. Disease-free survival (DFS) after primary tumor resection according to primary tumor location. FIG. 4. Disease-free survival (DFS) after primary tumor resection according to patient sex. Multivariate Analysis By multivariate analysis, we identified five independent factors that negatively influence DFS (Table 3): for model 1, MI >5/50 HPF (P =.004), primary tumor size >5 cm (P =.001), male sex (P =.003), R1 resection/tumor rupture (P =.04) and primary tumor location outside the stomach (P =.02); and for model 2, high/intermediate risk primary tumor (P <.0001 and P =.008, respectively), male sex (P =.007), R1 resection/tumor rupture (P =.01), and nongastric primary tumor location (P =.02). The initial model contained the following variables that were significant in the bivariate analysis: sex, MI (according to the NIH consensus categories), tumor size (according to the consensus categories), localization (gastric vs. other), and extent of surgery. Of the two-way interactions considered, the interaction of the categorized number of mitoses and the categorized tumor size was significant by partial likelihood test (P =.0115). The model is summarized in Table 3 (model 1). In model 2 (Table 3), the risk categories were entered instead of MI and tumor size in any form. Figure 5 shows the graphical assessment of the scale of numerical covariates of tumor size and number of mitoses. Curves provided by two residualbased methods are displayed along with the lines fitted by categorizing MI and tumor size. The significance of the deviation from the linear trend was assessed by modeling the curves with fractional polynomials. Both the residual based methods and the fractional polynomials method indicate that the tumor size (cm) can be entered into the model as a linear covariate without transformation. On the other hand, the examination of the functional form of the number of mitoses resulted in several transformations, producing similar curves that fit the data much better than the linear trend (the best transformation was a combination of 1/ÖIM, and ln(im))/öim) (model 3, data not shown). According to this analysis, an increase of 1 cm in tumor size is associated with recurrence risk increase of 7% (95% CI, 3 10). The curve for mitotic is consistent with the NIH categorization. In practice, both factors contribute similar risks for values of <10. However, considering the range of 10 to 20, the MI confers a greater risk increase. In contrast, for the numbers >20, risk is relatively stable for the MI, but still increases for tumor size. This suggests that large tumors with small numbers of mitoses have a worse prognosis than small tumors with a larger number of mitoses. Thus, it is worthwhile to emphasize that in common practical situations, the NIH risk categories (especially the high risk category) seem to underestimate the prognostic value of the MI. In all models, there are two additional risk factors that are consistently associated with worse prognosis: male sex and nongastric localization. Extent of surgery was significant in the models, including tumor size as a categorized variable. However, this is most probably the result of residual confounding associated with tumor size (a strong correlation between the R1 and the tumor size) because this variable becomes insignificant when tumor size is included as a numerical covariate in model 3. Although all three models have comparable R 2 values for goodness of fit (data not shown), the test for lack of fit is significant for model 2, suggesting that this model might not provide an appropriate fit. The histological type of the tumor was available for 207 patients. Adjusted hazard ratio for recurrence for spindle cell as opposed to other tumor types was.90 (P =.7209) and 1.02 (P =.9473) for extended models 1 and 3, respectively, thus not supporting the

7 2024 P. RUTKOWSKI ET AL. TABLE 3. Summary of multivariate analysis of factors associated with disease-free survival Model 1 (n = 274) Model 2 (n = 308) Variable P value HR HR 95% CI P value HR HR 95% CI Sex Male vs. female MI (/50 HPF) and TS (cm) MI 6 10, TS 5 vs. MI 5, TS MI>10, TS 5 vs. MI 5, TS MI 5, TS 6 10 vs. MI 5, TS MI 5, TS>10 vs. MI 5, TS MI 6 10, TS 6 10 vs. MI 5, TS 5 < MI >10, TS 6 10 vs. MI 5, TS 5 < MI 6 10, TS >10 vs. MI 5, TS MI >10, TS >10 vs. MI 5, TS Risk group HiR vs. LR < IR vs. LR Localization Gastric vs. other Extent of surgery R1 vs. R HR, hazard ratio; 95% CI, 95% confidence interval; MI, mitotic index; TS, tumor size; HiR, high risk, IR, intermediate risk; LR, low risk; R0, microscopically radical resection vs. R1, microscopically nonradical, but macroscopically radical resection or tumor intraoperative rupture. FIG. 5. Plots for assessing the scale of the mitotic index (MI) and tumor size as numerical variables. (A) Martingale residuals and lowest smoothed residuals from model. (B) f = log (smoothed censor/smoothed expected) + b VARIABLE, where b is the estimated coefficient for the variable of interest (i.e., tumor size on the upper panel and MI on the lower panel). (C) Estimated coefficients for the grouped tumor size and grouped MI ( 5, 6 10, 11 20, >20). hypothesis that the tumor type impacts the DFS prognosis. DISCUSSION Gastrointestinal stromal tumors (GISTs) have recently been morphologically, immunohistochemically, and molecularly distinguished from other gastrointestinal tract spindle cell neoplasms of mesenchymal origin. Radical surgery is the treatment of choice in primary resectable gastrointestinal stromal tumors, but virtually all GISTs are associated with a risk of recurrence, and approximately half of patients with potentially curative resections develop recurrent or metastatic disease Wide, multivisceral resections

8 PREDICTING RECURRENCE IN PRIMARY GIST 2025 with lymph node dissections are usually not necessary because of the low incidence of lymph node metastases, which was also observed in our series. Nevertheless, avoiding tumor rupture is imperative. The frequency of GIST diagnoses has greatly increased since 1992, along with a better understanding of the pathogenesis and biology of GIST and the routine availability of CD117 antigen immunostaining. 16 The survival in advanced cases has improved since 2000, when imatinib mesylate was introduced into clinical practice. 3,17 The objective of our study was to validate the importance of the NIH risk criteria and other prognostic factors for the outcome of treatment for primary GIST in a large, prospectively collected, nonselected group of patients in different locations that were included in the contemporary tumor registry, who had confirmed positive CD117 immunostaining. Most of the essential studies that determined the basis for the NIH risk criteria were published before the introduction of CD117 immunohistochemistry (a crucial test to confirm the contemporary diagnosis of GIST), and included both primary resectable and overtly malignant tumors, 11,12,14,18 21 or were based on a small number of cases. 22,23 To our knowledge, the value of the NIH criteria for primary tumors has not been confirmed in a large homogenous series of primary resectable GISTs. Miettinen et al analyzed the value of risk factors according to the anatomical site of the primary tumor and created their own classifications for risk assessment in gastric, duodenal, and intestinal GISTs. On the other hand, adjuvant trials after resection of primary GISTs (e.g., European Organization for Research and Treatment of Cancer trial or American College of the Surgeons Oncology Group trial Z ) and the Scandinavian study 8,27 29 on retrospective population data called into question the poorer prognosis in the intermediate risk group, thus raising debate regarding the accuracy of the risk criteria in primary GISTs. Overall, DFS in the group of nonselected GIST patients in this study was 38% at 5 years, which is similar to results observed in other published series. 11,20,30,31 However, this disease represents a broad spectrum of different biological behaviors. Complete resection is curative for most low and very low risk patients, and this group is characterized by an excellent prognosis. In contrast, the prognosis for the high risk group is poor, and surgery alone is insufficient for high-grade tumors. Our study confirms the importance of stratification of the intermediate risk group, the existence of which was criticized in a recent population-based study. 8 In our opinion, patients with an intermediate risk of recurrence should also be recommended for adjuvant trials with small molecule inhibitors, such as imatinib, because the risk of recurrence is relatively high after long-term followup. Therefore, the stratification of risk categories (intermediate and high) is warranted in such trials. The impact of classical prognostic factors on DFS in the current study is consistent with the results of other studies. 11,14,30,32 Thus, we confirm the role of MI and tumor size as the most valuable classic prognostic factors for primary GISTs, as included in the NIH risk criteria. 4 However, our data suggest that the value of the MI is underestimated for tumors of <20 cm in the largest dimension, and should be considered as the most important variable. On the other hand, tumor size >20 cm predicts the greatest risk. Moreover, the estimation of tumor size gives the best prognostic prediction when considered as a continuous variable, and in the case of larger tumors, the value of microscopic radicalism of surgical treatment diminishes. We also confirm the effect of the anatomic site of the primary lesion on the prognosis in GIST, which was previously suggested by other studies. 3,30,33 Primary tumors in the gastric area have a far better prognosis those that in a nongastric location, independent of tumor size and MI. It is unknown, however, the degree to which this may be because tumors outside the stomach can growth asymptomatically for a longer time. Nevertheless, the stomach was the most common primary site of tumors in our series, whereas the small intestine was the most frequent primary location in the group of patients who experienced recurrence. No statistically significant correlation between mutation status and the disease recurrence after surgical treatment of primary GISTs was found, even in subgroups with common exon 11 KIT mutations, which were present in two-thirds of the cases in our series, or exon 11 missense mutations versus deletions/insertions, as suggested by some researchers 31,34 36 despite some disclaimers by others We cannot neglect the bias related to the small number of cases with confirmed mutational status; however, KIT mutations are acquired early in GIST development, so they can occur in low and very low risk tumors. We did not analyze the role of DNA ploidy, immunohistochemical staining for cell proliferation antigens, or other factors involved in regulation of the cell cycle (p53, Ki-67, p21, or BCL-2) indicated by other authors as important prognostic factors This was because our goal was to establish a set of

9 2026 P. RUTKOWSKI ET AL. factors that could be used in routine practice. Other authors also indicated the significance of factors such mucosal invasion or tumor necrosis as indicators for risk of malignant behavior, but these factors are poorly reproducible. GIST recurrence can be observed late after treatment for the primary tumor, which indicates the necessity for a longer follow-up than the 5 years usually set as the cutoff point for most patients with sarcoma. Low malignant potential GIST should be followed up on a yearly basis because recurrence is rare in this group of patients. However, patients should be informed about the likelihood of late relapse, because no form of GIST can be labeled truly benign. In higher risk patients, recurrences will occur continuously for years after surgical resection (Fig. 2). Because symptomatology does not provide early diagnostic clues, follow-up with computed tomography imaging of the abdomen and pelvis (because recurrences typically occur first in the liver or peritoneum 2,3 ) is recommended every 3 to 6 months for at least 5 years in high and intermediate risk groups. In summary, although the NIH criteria provide an excellent estimation of tumor behavior, physicians should also be familiar with other factors that influence a GIST patientõs prognosis. Our data from a large series of patients undergoing resection of GIST tumors confirm that the NIH risk criteria (which, as shown in Fig. 2, perform quite well to categorize risk of recurrence), with slightly more emphasis on MI as a discrete variable and on tumor size as a continuous variable. Additional prognostic factors identified here were tumor location (gastric vs. other) and patient sex. All identified or confirmed prognostic variables are simple and easily applicable in the everyday practices of physicians treating patients with GIST. ACKNOWLEDGMENTS Supported by the Polish State Committee for Scientific Research, grant 3P05C We thank all doctors devoted to GIST treatment and diagnosis who collaborated in the Polish Clinical GIST Registry, M. Rosinska for statistical advice, and J. Lasota (Armed Forces Institute of Pathology, Washington, DC) for critical remarks and mutational analyses data. REFERENCES 1. Kindblom LG, Remotti HE, Aldenborg F, Meis-Kinblom JM. Gastrointestinal pacemaker cell tumor (GIPACT): gastrointestinal stromal tumors show phenotypic characteristic of the interstitial cell of Cajal. Am J Pathol 1998; 152: Miettinen M, Lasota J. Gastrointestinal stromal tumors definition, clinical, histological, immunohistochemical, and molecular genetic features and differential diagnosis. Virchows Arch 2001; 438: Trent JC, Benjamin RS. New developments in gastrointestinal stromal tumor. Curr Opin Oncol 2006; 18: Fletcher CDM, Berman JJ, Corless C, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 2002; 33: Miettinen M, Majidi M, Lasota J. Pathology and diagnostic criteria of gastrointestinal stromal tumors (GISTs): a review. Eur J Cancer 2002; 38(Suppl 5):S Miettinen M, El-Rifai W, Sobin LH, Lasota J. Evaluation of malignancy and prognosis of gastrointestinal tumors: a review. Hum Pathol 2002; 33: Eisenberg BL, Judson I. Surgery, imatinib in the management of GIST: emerging approaches to adjuvant and neoadjuvant therapy. Ann Surg Oncol 2004; 11: Nilsson B, Bumming P, Meis-Kindblom JM, et al. Gastrointestinal stromal tumors: the incidence, prevalence, clinical course, and prognostication in pre-imatinib mesylate era. Cancer 2005; 103: Debiec-Rychter M, Dumez H, et al. EORTC Soft Tissue and Bone Sarcoma Group. Use of c-kit/pdgfra mutational analysis to predict the clinical response to imatinib in patients with advanced gastrointestinal stromal tumours entered on phase I and II studies of the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer 2004; 40: Hosmer DW, Lemeshow S. Applied Survival Analysis: Regression Modeling of Time to Event Data New York: J Wiley and Sons; DeMatteo RP, Lewis JJ, Leung D, Mudan SS, Woodruff JM, Brennan MF. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000; 231: Kim CJ, Day S, Yeh KA. Gastrointestinal stromal tumors: analysis of clinical and pathologic factors. Am Surg 2001; 67: Hinz S, Pauser U, Egberts JH, Schafmayer C, Tepel J, Fandrich F. Audit of a series of 40 gastrointestinal stromal tumour cases. Eur J Surg Oncol 2006; 32: Pierie JP, Choudry U, Muzikansky A, et al. The effect of surgery and grade on outcome of gastrointestinal stromal tumors. Arch Surg 2001; 136: Ng E, Pollock RE, Munsell MF, et al. Prognostic factors influencing survival in gastrointestinal leiomyosarcomas. Ann Surg 1992; 215: Perez EA, Livingstone AS, Franceschi D, et al. Current incidence and outcomes of gastrointestinal mesenchymal tumors including gastrointestinal stromal tumors. J Am Coll Surg 2006; 202: Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002; 347: Bucher P, Egger JF, Gervaz P, et al. An audit of surgical management of gastrointestinal stromal tumours (GIST). Eur J Surg Oncol 2006; 32: Clary BM, Dematteo RP, Lewis JJ, et al. Gastrointestinal stromal tumors and leiomyosarcomas of the abdomen and retroperitoneum: a clinical comparison. Ann Surg Oncol 2001; 8: Crosby JA, Catton CN, Davis A, et al. Malignant gastrointestinal stromal tumors of the small intestine: a review of 50 cases from a prospective database. Ann Surg Oncol 2001; 8: Miettinen M, Furlong M, Sarlomo-Rikala M, Burke A, Sobin LH, Lasota J. Gastrointestinal stromal tumors, intramural leiomyomas, and leiomyosarcomas in the rectum and anus. A

10 PREDICTING RECURRENCE IN PRIMARY GIST 2027 clinicopathologic, immunohistochemical, and molecular genetic study of 144 cases. Am J Surg Pathol 2001; 25: Aparicio T, Boige V, Sabourin JC, et al. Prognostic factors after surgery of primary resectable gastrointestinal stromal tumours. Eur J Surg Oncol. 2004; 30: Bucher P, Taylor S, Villiger P, Morel P, Brundler MA. Are there any prognostic factors for small intestinal stromal tumors?. Am J Surg 2004; 187: Miettinen M, Sobin LH, Lasota J. Gastrointestinal stromal tumors of the stomach: a clinicopathologic, immunohistochemical, and molecular genetic study of 1765 cases with longterm follow-up. Am J Surg Pathol 2005; 29: Miettinen M, Kopczynski J, Makhlouf HR, et al. Gastrointestinal stromal tumors, intramural leiomyomas, and leiomyosarcomas in the duodenum: a clinicopathologic, immunohistochemical, and molecular genetic study of 167 cases. Am J Surg Pathol. 2003; 27: Miettinen M, Makhlouf H, Sobin LH, Lasota J. Gastrointestinal stromal tumors of the jejunum and ileum: a clinicopathologic, immunohistochemical, and molecular genetic study of 906 cases before imatinib with long-term follow-up. Am J Surg Pathol 2006; 30: Kindblom LG, Meis KindBlom J, Bumming P, et al. Incidence, prevalence, phenotype and biologic spectrum of gastrointestinal stromal tumors (GIST): a population based study of 600 cases. Ann Oncol 2003; 13: Buemming P, Meis-Kindblom JM, Kindblom LG, et al. Is there an indication for adjuvant treatment with imatinib mesylate in patients with aggressive gastrointestinal stromal tumors (GISTs)? (abstract 3289). Proc Am Soc Clin Oncol 2003; 22: Bumming P, Ahlman H, Andersson J, Meis-Kindblom JM, Kindblom LG, Nilsson B. Population-based study of the diagnosis and treatment of gastrointestinal stromal tumours. Br J Surg 2006; 93: Emory TS, Sobin LH, Lukes L, Lee DH, OÕLeary TJ. Prognosis of gastrointestinal smooth-muscle (stromal) tumors: dependence on anatomic site. Am J Surg Pathol 1999; 23: Singer S, Rubin BP, Lux ML, et al. Prognostic value of KIT mutation type, mitotic activity and histologic subtype in gastrointestinal stromal tumors. J Clin Oncol 2002; 20: Fujimoto Y, Nakanishi Y, Yoshimura K, Shimoda T. Clinicopathologic study of primary malignant gastrointestinal stromal tumor of the stomach, with special reference to prognostic factors: analysis of results in 140 surgically resected patients. Gastric Cancer 2003; 6: Ueyama T, Guo KJ, Hashimoto H, et al. A clinicopathologic and immunohistochemical study of gastrointestinal stromal tumors. Cancer 1992; 69: Andersson J, Bumming P, Meis-Kindblom JM, et al. Gastrointestinal stromal tumors with KIT exon 11 deletions are associated with poor prognosis. Gastroenterology 2006; 130: Kim TW, Lee H, Kang YK, et al. Prognostic significance of c- kit mutation in localized gastrointestinal stromal tumors. Clin Cancer Res 2004; 10: Taniguchi M, Nishida T, Hirota S, et al. Effect of c-kit mutation on prognosis of gastrointestinal stromal tumors. Cancer Res 1999; 59: Corless CL, McGreevey L, Haley A, Twon A, Heinrich MC. KIT mutations are common in incidental gastrointestinal stromal tumors one centimeter or less in size. Am J Pathol 2002; 160: Corless CL, Fletcher JA, Heinrich MC. Biology of gastrointestinal stromal tumors. J Clin Oncol 2004; 22: Lasota J, Miettinen M. KIT and PDGFRA mutations in gastrointestinal stromal tumors (GISTs). Sem Diagn Pathol 2006; 23: Carillo R, Candia A, Rodriguez-Peralto JL, Caz V. Prognostic significance of DNA ploidy and proliferative index (MIB-1 index) in gastrointestinal stromal tumors. Hum Pathol 1997; 28: Ozguc H, Yilmazlar T, Yerci O, et al. Analysis of prognostic and immunohistochemical factors in gastrointestinal stromal tumors with malignant potential. J Gastrointest Surg 2005; 9: Cunningham RE, Federspiel BH, McCarthy WF, Sobin LH, OÕLeary TJ. Predicting prognosis of gastrointestinal smooth muscle tumors. Role of clinical and histologic evaluation, flow cytometry, and image cytometry. Am J Surg Pathol 1993; 17: Liu FY, Qi JP, Xu FL, Wu AP. Clinicopathological and immunohistochemical analysis of gastrointestinal stromal tumor. World J Gastroenterol 2006; 12: Hu TH, Chuah SK, Lin JW, et al. Expression and prognostic role of molecular markers in 99 KIT-positive gastric stromal tumors in Taiwanese. World J Gastroenterol 2006; 12: Wong NA, Young R, Malcomson RD, et al. Prognostic indicators for gastrointestinal stromal tumours: a clinicopathological and immunohistochemical study of 108 resected cases of the stomach. Histopathology 2003; 43: