Improved Efficacy of First-Line Imatinib in Advanced Gastrointestinal Stromal Tumors (GIST): The Dutch GIST Registry Data

Gastrointestinal stromal tumors (GISTs) represent the most common mesenchymal neoplasm of the gastrointestinal tract, affecting 17 patients per million per year [1]. Most primary GISTs are found in the stomach and the small intestine, while the remaining minority is located at other sites of the gastrointestinal tract (e.g. esophagus, colon, and rectum). The majority of cases present with localized disease, however about 15% of patients have metastatic disease at presentation [2]. Furthermore, 5 years after complete surgical removal of GIST, 30% of patients will have recurrence or metastases [3].

The introduction of imatinib, a tyrosine-kinase inhibitor with activity against BCR-ABL, KIT, and PDGFRA receptors, led to greatly improved prognosis of patients with advanced GIST. In early 2000s, two phase III trials [4, 5], including the 62005 trial conducted by the European Organization for Research and Treatment of Cancer (EORTC), demonstrated the efficacy of imatinib in the treatment of unresectable and metastatic GIST [4]. Since then, imatinib became the indisputable first-line treatment in metastasized and unresectable GIST. With a median follow-up of over 10 years, long-term results demonstrated a median progression-free survival (PFS) of 1.7–2.0 years in patients receiving imatinib, with an estimated PFS at 10 years of 9.2–9.5% [6].

Now, almost 2 decades after the introduction of imatinib, it is of interest for patients, clinicians, and researchers to know the current outcomes of imatinib treatment. In other soft tissue sarcoma’s, it has been reported that outcomes can improve a.o. due to better supportive care [7, 8]. To explore this in GIST, we compared the clinical outcomes in a large and recent patient cohort with clinical outcomes of patient populations in early phase III trials.

Overall, 435 patients with advanced GIST registered in the DGR were treated with imatinib as first-line palliative treatment. Of these patients, 420 had recorded response evaluation and could be included in the analysis (Fig. 1). Fifteen patients (3.4%) were lost to follow-up and were not included in the analysis. Demographic and clinical features are listed in Table 1. The starting dose of imatinib was 400 mg daily, except for patients with a KIT-exon 9 mutation who were treated with 800 mg daily. Dose reduction (due to intolerance) was observed in 59 (14.0%) patients. Six patients underwent metastasectomy (liver, n = 2; peritoneal, n = 4) in addition to systemic treatment. In the majority of patients (n = 328, 78.1%), the first systemic treatment received was palliative imatinib, while the remainder of the patients (n = 92, 21.9%) had a history of treatment with (neo)adjuvant imatinib.

The median follow-up duration was 35.0 months (range 2.0–136.0). Treatment of first-line imatinib resulted in complete or partial response as best response in 238/420 patients: ORR 56.7% (Table 2). During the follow-up, 217 patients (51.2%) eventually showed progression of disease. The median PFS in our cohort was 33.0 months (95% CI 28.4–37.6) (Fig. 2).

After treatment with imatinib, 180 patients were treated with sunitinib, with an ORR of 19%. In 82 patients receiving third-line therapy with regorafenib, an ORR of 14% was observed.

Exploring the survival data of the DGR revealed that 159/420 (37.9%) patients died during the follow-up. The causes of death were progression of GIST in 132 patients, other malignancies (rectal cancer, lung cancer, metastatic melanoma, leukemia, and adenocarcinoma of the stomach) in six patients, and non-malignant diseases (cardiovascular diseases, sepsis, hepatic failure) in 10 patients. An unspecified cause of death was reported in 9 patients. A median OS of 68.0 months (95% CI 56.1–80.0) was observed in patients with first-line imatinib, and the OS estimates at 1 and 2 years were 94% and 84%, respectively (Fig. 3).

Patients with a KIT-exon 11 mutation had a median PFS of 38.0 months (95% CI 30.0–46.0), while a median PFS of 25 months (95% CI 7.4–42.6) was observed in patients having a KIT-exon 9 mutation (p = 0.034). OS analysis showed the same trend; patients with a KIT-exon 11 had a longer OS (73.0 months, 95% CI 57.1–88.9) compared with KIT-exon 9 patients (57 months, 95% CI 48.0–66.0; p = 0.042). The Kaplan–Meier curves for PFS and OS for both KIT-exon 11 and KIT-exon 9 are shown in Figs. 4 and 5.

Studying potential prognostic variables for PFS using Cox regression multivariable analysis (Table 3) showed that sex, age, performance status, location of the primary GIST, and sum of target lesions were not significant prognostic factors. Although just not significant, patients with wild-type KIT/PDGFGR/SDH/BRAF GIST had shorter PFS.

There is no doubt that imatinib revolutionarily improved the prognosis of GIST patients after confirmation of its efficacy in clinical trials. In a US/Finland phase II trial [11] with a median follow-up duration of 63 months, patients taking imatinib (400 or 800 mg) had an overall median time to progression of 24 months, with a median OS of 57 months. The phase III S0033 trial [5] demonstrated a median PFS of 18 months and median OS of 55 months in patients receiving imatinib 400 mg/day. In EORTC 62055, imatinib 400 mg/day led to an estimated median PFS of 19.5 months.

In the 62005 EORTC phase III trial [4], having a median follow-up duration of 760 days (25.3 months), 473 patients received 400 mg once daily and 473 patients were treated with high-dose imatinib of 800 mg (400 mg twice daily). In patients who were allocated to a daily dose of 400 mg, the proportion of patients with a complete response (n = 24) or partial response as best response (n = 213) was 52.9%. Progression of disease occurred in 263 patients (56%) assigned to imatinib 400 mg and 235 patients (50%) treated with imatinib 800 mg. In the published results of the 62005 trial, the exact duration of PFS was not mentioned in the article; however, when assessing the provided Kaplan–Meier curves, a PFS of 19.5 months can be estimated for patients treated with imatinib 400 mg once daily [4].

In our study, we observed a considerably longer PFS (33 months) than in EORTC 62005 and the other two trials, which were the trials resulting in approval of imatinib 400 mg/day as first-line treatment of advanced GIST and marked the beginning of the era of imatinib in GIST. Furthermore, a higher proportion of patients in the DGR had partial or complete response as best response compared with the 62005 and S0033 trials (Table 4). While median OS in the EORTC 62005 trial was not reached at the time of the published article in 2004, long-term results [6] reported an OS of 3.9 years (46.8 months) among patients treated with imatinib 400 mg. In our cohort, the median OS of 5.7 years (68.0 months) is markedly longer than the introduction time of imatinib in the early 2000s.

A major reason for the superiority of clinical outcomes in our study compared with the EORTC trial and other clinical trials is probably patient selection. Before the introduction of imatinib, no effective treatment was available for advanced GIST (GIST is unresponsive to conventional chemotherapy) [12] and therefore patients participating in the early phase III trials (e.g. EORTC 62005) were mainly those with metastatic GIST who had multiple voluminous lesions. It is very likely that these patients had poorer prognosis than patients treated nowadays (e.g. patients in the DGR), as we know from previously published data that a high tumor burden is a negative prognostic factor in metastatic GIST [13, 14]. Due to advancements in recognizing and diagnosing GIST and timely starting of imatinib, patients are treated when tumor burden is lower. Furthermore, acquired experience in managing the adverse events of imatinib, and the availability of second-, third- and fourth-line therapy, have led to improved prognosis of patients with advanced GIST since the initial phase of the introduction of imatinib.

Better outcomes of treatment with established tyrosine kinase inhibitors (TKIs; the ‘control arm’) for GIST have been reported in several recent clinical trials compared with older data from registration trials. For example, in the Intrigue trial comparing the efficacy of ripretinib (new TKI) with sunitinib (established second-line) as second-line therapy of advanced GIST, patients treated with sunitinib had a median PFS of 8.3 months [15], which is longer than the observed PFS in the registration trial (PFS of 5.6 months) [16]. The same trend applies for the Voyager trial, in which avapritinib was compared with regorafenib in GIST patients who did not respond to prior treatment with imatinib and sunitinib. Patients treated with regorafenib showed a median PFS of 5.6 months as third-line therapy in advanced GIST [17], while in the registration trial, a median PFS of 4.8 months was observed [18].

In the current study, nearly two decades after the introduction of imatinib, we present an update on the outcomes of treatment with imatinib in patients with unresectable and metastasized GIST. The data were retrieved from a real-life database, including clinical details of GIST patients treated in five sarcoma specialized centers in The Netherlands. The results of this study are a fair representation of the outcomes of treatment of GIST patients in a real-world setting. Therefore, these outcomes could serve as an updated reference model when outcomes of (new) agents are compared with imatinib.

The limitations of this study were the retrospective study design and the absence of response status in a proportion (3.4%) of patients, which may have biased the results. Nevertheless, our study contains detailed information on clinical features and outcomes of a relatively large cohort of GIST patients with advanced disease.

This study presents an update of the efficacy of imatinib in the treatment of advanced GIST, and demonstrates that clinical outcomes of first-line imatinib are improved compared with the beginning era of imatinib in GIST. Furthermore, these results represent outcomes in real-world clinical practice and can serve as a reference for what can be expected from imatinib in the first-line setting in advanced GIST.