The outcome and predictive factors of sunitinib therapy in advanced gastrointestinal stromal tumors (GIST) after imatinib failure - one institution study

Background Gastrointestinal stromal tumors (GIST) mutational status is recognized factor related to the results of tyrosine kinase inhibitors therapy such as imatinib (IM) or sunitinib (SU). Arterial hypertension (AH) is common adverse event related to SU, reported as predictive factor in renal cell carcinoma. The aim of the study was to analyze the outcomes and factors predicting results of SU therapy in inoperable/metastatic CD117(+) GIST patients after IM failure. Methods We identified 137 consecutive patients with advanced inoperable/metastatic GIST treated in one center with SU (2nd line treatment). Median follow-up time was 23 months. Additionally, in 39 patients there were analyzed selected constitutive single nucleotide polymorphisms (SNPs) of VEGFA and VEGFR2 genes. Results One year progression-free survival (PFS; calculated from the start of SU) rate was 42% and median PFS was 43 weeks. The estimated overall survival (OS, calculated both from start of SU or IM) was 74 weeks and 51 months, respectively. One-year PFS was 65% (median 74 weeks) in 55 patients with AH vs. 22% (median 17 weeks) in patients without AH. Patients with primary tumors carrying mutations in KIT exon 9 or wild-type had substantially better 1-year PFS (68% and 57%; median 65.5 and 50.5 weeks, respectively) than patients having tumors with KIT exon 11 or PDGFRA mutations (34% and 15%; median 36.8 and 9 weeks, respectively). We identified two independent factors with significant impact on PFS and OS in univariate and multivariate analysis: primary tumor genotype and presence of AH. The most common adverse events during therapy were: fatigue, AH, hypothyroidism, hand and foot syndrome, mucositis, skin reactions, dyspepsia, and diarrhea. Two deaths were assessed as related to tumor rupture caused by reaction to SU therapy. The presence of C-allele in rs833061 and the T-allele in rs3025039 polymorphism of VEGFA were associated with significantly higher risk of hypothyroidism (OR: 10.0 p = 0.041 and OR: 10.5; p = 0.015, respectively). Conclusions We confirmed that many advanced GIST patients benefit from SU therapy with OS > 1.5 year. Primary tumor KIT/PDGFRA genotype and SU-induced AH, as surrogate of its antiangiogenic activity are two independent factors influencing both PFS and OS. Note The preliminary data of this study were presented during Annual Meeting of American Society of Clinical Oncology, 4-8 June 2011 and Connective Tissue Oncology Society Meeting, 26-28 October 2011 in Chicago, IL.


Background
Unprecedented improvement in advanced gastrointestinal stromal tumors (GIST management has been achieved due to recent recognition of the important biological role of activating mutations in KIT and PDGFRA (platelet-derived growth factor receptor-alpha) genes. Those observations led to the introduction of imatinib mesylate, a small-molecule selective inhibitor of the receptor tyrosine kinases such as stem-cell factor receptor (KIT, CD117), BCR-ABL and platelet-derived growth factor receptors (PDGFRs)-A and -B. Imatinib revolutionized the outcome of patients with advanced CD117positive GISTs and is currently approved as the first-line treatment in advanced (metastatic and/or inoperable) GISTs [1][2][3][4][5]. However, the spectacular response to imatinib therapy is time-limited and secondary resistance to imatinib therapy (after initial stabilization or response) develops in majority of patients [4].
Currently, the only approved second-line drug is sunitinib malate -a multitargeted agent, an inhibitor of tyrosine kinase, of KIT and PDGFRA/B and of the vascular endothelial growth factor receptors (VEGFRs)-1, -2 and 3, FMS-like tyrosine kinase-3 (FLT3), colony stimulating factor 1 receptor (CSF-1R), and glial cell-line derived neurotrophic factor receptor (REarranged during Transfection; RET) [6][7][8][9][10][11]. Sunitinib possesses both antiangiogenic and cytostatic properties and by competing with ATP binding prevents multiple receptor tyrosine kinases phosphorylation in vitro and in vivo. Two phase II, one phase III and one "treatment-use" trials have investigated the activity of sunitinib in GIST patients after the failure of prior imatinib treatment, and all these trials have shown the significant activity of sunitinib in this population of patients [11][12][13][14]. The objective clinical benefit was achieved in approximately 60% of GIST patients who received sunitinib after failure of prior imatinib treatment [11][12][13][14]. Median progression -free survival time on sunitinib is 6-8 months. The adverse events reported during this therapy are frequent. The most common treatment-related adverse events were fatigue, diarrhea, skin discoloration, nausea, mucositis, arterial hypertension, hand and foot syndrome (palmarplantar erythrodysesthesia), impairment of left ventricular ejection fraction and hypothyroidism [12,14].
Moreover, arterial hypertension was not only the common adverse event during sunitinib therapy, but it was reported as predictive factor for results of renal-cell carcinoma (RCC) patients [15,16]. This phenomenon has not been yet analyzed in GIST patients.
There is a lack of studies analyzing the outcome of sunitinib in advanced GISTs after imatinib failure therapy in routine practice outside clinical trials. Thus, the aim of our study was to evaluate factors predicting results and toxicity of SU second-line therapy in inoperable/metastatic GISTs. Additionally, we have investigated the impact of the selected single nucleotide polymorphisms (SNPs) in VEGFA and VEGFR2 genes on sunitinib-related toxicity in the subgroup of patients.

Patients
We analyzed prospectively collected data of 137 consecutive patients treated with sunitinib maleate because of inoperable and/or metastatic CD117 positive GIST enrolled into therapy between October, 2005 and February, 2011, reviewed in one tertiary cancer center. All patients met the following criteria for sunitinib treatment: 1) histological diagnosis of GIST, confirmed by CD117-immunopositivity (DAKO; Carpintiera, CA), 2) metastatic and/or inoperable lesions after failure on prior treatment with imatinib (confirmed progressive disease or unacceptable toxicity) 3) measurable disease on computed tomography (CT) scans, 4) WHO performance status ≤3, 5) no concomitant therapy for disease, 6) adequate renal, cardiac and liver function.
Each patient provided informed consent for the study and collection of clinical and molecular data prior sunitinib therapy. The study had been approved by the local Bio-Ethics Committee according to Good Clinical Practice Guidelines (approvals from Bio-Ethics Committees from Medical University of Gdansk and from Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Warsaw KB/9/2011 and approval for Polish Clinical GIST Registry by Internal Review Board 119/2002). Patients provided additional informed consent for taking the 5 ml blood samples for gene polymorphisms analysis. Patients did not undergo any further selection; 35 patients were initially included in the treatment-use trial A6181036.
All patients were treated with sunitinib in initial licensed dose of 50 mg daily in 6 weeks cycle (4 weeks on/2 weeks off therapy), however the dosing could be reduced (to 37.5 mg or 25 mg) or delayed or modulated to the dosing of 37.5 mg on continuous schedule to optimize the benefit-risk profile according to decision of treating physician. The treatment was continued until confirmed progression of the disease or unacceptable toxicity. All patients were followed carefully with median follow-up time of 23 months (range: 6-68 months). The objective response of GIST to sunitinib therapy was evaluated with serial CT examinations (performed every 2-3 months), according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.0 [17]. In case of progression, patients were treated with other different tyrosine kinase inhibitors or cytotoxic chemotherapy or best supportive care only. If possible, they were included into clinical trials with new compounds. Toxic effects were graded with National Cancer Institute common toxicity criteria, version 3.0 [18].

Genotyping
Genomic screening was performed for the presence of the KIT (exons 9, 11, 13, and 17) or PDGFRA (exons 12, 14 and 18) genes mutation in randomly selected 89 cases, based on DNA isolated from paraffin-embedded or fresh frozen imatinib-naive tumor tissues, as previously described [19].

SNPs analysis
Genomic DNA was isolated from peripheral blood samples of 39 consenting patients using the standard protocol with proteinase K digestion, phenol -chloroform extraction and ethanol precipitation. Three selected VEGFA SNPs: rs699947 (-2578 C > A), rs3025039 (+936 C > T), and rs2010963 (+405 G > C) and two VEGFR2 (KDR) SNPs: rs1531289 (3405-92A > G) and rs1870377 (+1416 T > A) were genotyped using restriction fragment length polymorphism (RFLP) method. The restriction enzymes which detected -SNPs mentioned above were BsaIII, BsmF1, NlaIII and AluI, respectively. Amplified DNA was digested with endonucleases overnight at optimal temperatures according to the manufacturer (Fermentas, Thermo Scientific) and then electrophoresed on 2% agarose gel. For identification of an additional VEGFA SNP: -rs833061 (460 T > C) direct sequencing was performed using ABI 3100 Genetic Analyser (Applied Biosystems Inc. Foster City, CA). Primers sequences and PCR cycling parameters are available on request.

Statistical analysis
Contingency tables were analyzed by the chi-square test. Progression-free survival (PFS) time was calculated from the date of the start of sunitinib treatment to the date of the most recent follow-up, or progression or death due to the disease. Overall survival (OS) time was calculated either from the date of the start of imatinib and sunitinib treatment to the date of the most recent follow-up or death due to the disease [except cases of two patients, whose death was attributed to adverse event (AE) of sunitinib therapy]. PFS was assessed with respect to the following variables: demographic data (age at the diagnosis ≤ 45 or > 45 years; gender), primary tumor genotype (KIT 11 exon, KIT 9 exon, any PDGFRA mutations and wild-type cases), length of previous therapy on imatinib (≤ 6, > 6-12, > 12 months), and presence of arterial hypertension (defined as occurrence of systolic blood pressure > 140 mm Hg or diastolic blood pressure > 90 mm Hg, or deterioration of preexisting AH during the first three months of therapy with sunitinib). The Kaplan-Meier method was used for analysis of survival curves, compared by log-rank test.
For univariate comparison of the survival between groups, the Kaplan-Meier estimator was used with generalized Wilcoxon and the log-rank tests. For PFS/OS comparisons in relation to presence of arterial hypertension two cases with early death due to tumor perforation were excluded to minimize the lead-time bias of the results. In multivariate analysis of the factors associated with PFS and OS, we used Cox proportional hazards models, applying the stepwise model building procedure that included all covariates significant at 20% level in bivariate analysis. Two-way interactions were then considered in the model. Differences were considered statistically significant if p-values were < 0.05. These statistical computations were performed using Statistica 6.1 software [Statsoft ® ; Tulsa, OK].
The Hardy-Weinberg equilibrium analysis of VEGFA and VEGFR2 polymorphisms was performed using a chi-square test with one degree of freedom. Multiplicative, dominant and recessive models were tested in every SNP for the associations with OS, PFS and treatment AEs. Odds ratios (ORs) with their exact 95% confidence intervals (CI) were calculated and the exact test for OR was used. The polymorphisms analyses were performed using StatXact-3 software, version 3.1.

Clinicopathological data
The distribution of clinical and pathological data of patients included in the study is listed in Table 1. There were 74 male and 63 female patients, with median age at the start of sunitinib therapy 55 years (range: 15 -82). The majority of primary tumors were located in the small intestine (57.6%), followed by the stomach (33.5%). All but two patients had documented progression on imatinib. Majority of patients (70.1%) were pre-treated with imatinib for more than one year. Almost 90% of patients started sunitinib therapy being in relatively good performance status (0-1).

Mutational analysis data
The distribution of patients according to the initial tumor mutational status is shown in Table 1. In the group of 89 patients, whose initial tumor mutational status was evaluated, 58.4% of GISTs had an exon 11 KIT mutation, 16.9% had an exon 9 KIT mutation, 13.5% had PDGFRA gene mutation (11 of 12 cases had D842V mutation) and in 11.2% of tumors we have not detected any mutations (wild-type).

Outcomes of sunitinib treatment
Median PFS was 43 weeks and estimated 1-year PFS rate was 42%. Progression of disease during sunitinib therapy was observed in 105 cases (77%). At the time of the analysis, 45 patients (33%) were alive. Estimated 2year OS rate was 40% and median OS was 73.5 weeks. Estimated 5-year OS in this group of patients was 45% (when calculated from the date of imatinib start) and median OS -51 months.
The best responses observed during sunitinib therapy and estimated by CT imaging (two consecutive examinations) according to RECIST criteria were as follows: none complete responses (CRs), 21 (15%) partial responses (PRs), 62 (45%) stable disease (SD) at least four months, 51 (37%) progressive disease (PD) and three patients were not assessable for response. Overall clinical benefit of sunitinib therapy (counted as the sum of CR, PR and SD rates) was 60%.

Factors influencing PFS and OS during sunitinib therapy
In univariate analysis two factors significantly correlated with shorter PFS and OS that were: tumor genotype: exon 11 KIT or PDGFRA mutation (p = 0.04 and p =  0.04, respectively) (Figure 1), and absence of arterial hypertension during sunitinib therapy (p = 0.0001 and p = 0.001) (Figure 2). We did not find any significant correlation between PFS/OS and patients gender and age at diagnosis, GIST primary tumor location, WHO performance status 0 vs. In the multivariate analysis two factors (tumor genotype and arterial hypertension) had independent predictive value for PSF and OS (Tables 3 and 4).

Genetic variations of VEGFA and VEGFR2 genes
The genotype frequency of VEGFA/VEGFR2 gene SNPs are presented in Table 5. All analyzed genotypes were in Hardy-Weinberg equilibrium. The higher risk of hypothyroidism during sunitinib therapy was associated with: C-allele in rs833061 (OR: 10.l; p = 0.041) and Tallele in rs3025039 (OR: 10.5; p = 0.015) ( Table 6). We did not find the correlation between the presence of SNPs and arterial hypertension, hand-foot-syndrome, skin toxicity/mucositis or diarrhea (data not shown).

Discussion
A majority of patients with advanced GISTs ultimately stop responding to imatinib and unquestionably management of disease resistant to first-line treatment represents a clinical challenge [4]. Insights into resistance mechanisms have allowed developing several strategies in patients with progression during imatinib treatment. In case of generalized progression (or intolerance to imatinib) the main option is using monotherapy with alternative multi-tyrosine kinase inhibitor -sunitinib, which remains the only approved second line drug for the treatment of advanced GISTs after imatinib therapy failure [20]. Sunitinib has demonstrated robust clinical effectiveness in imatinib-resistant or -intolerant GIST as shown in randomized, placebo-controlled phase III trial in which the median time to tumor progression for patients treated with sunitinib was more than four times longer than that for patients receiving placebo (27.3 vs. 6.4 weeks) [12]. Present study, according to our best knowledge, represents of the largest series of GIST patients after imatinib failure analyzed for the outcome of sunitinib treatment in routine clinical practice outside randomized, controlled clinical trial.
We have also attempted to prove tumor genotype implications and to find new predictive factors in this group of patients. We have confirmed that many   advanced GIST patients benefit from sunitinib therapy (mainly due to stabilization of disease according to RECIST, not Choi criteria [21]) with OS exceeding 1.5 years. The median PFS longer than seven months is almost equal to the results of the Korean one-institution study [22]. We have also confirmed in more detailed way and on the larger group of patients, than ever published, data regarding the correlation between primary tumor mutational status and sunitinib treatment outcomes [22][23][24]. As for imatinib, KIT mutation status appears to serve as a predictor of tumor response to sunitinib. We have proven that, contrary to imatinib, tumors initially (pre-imatinib treatment) bearing KIT exon 9 mutation or with wild-type genotype have a higher chance to respond to sunitinib. Moreover, GISTs harboring KIT exon 9 mutations appear to be more sensitive to sunitinib than those with primary KIT exon 11 mutations (however we have observed some objective responses also in this group of patients). The clinical benefit of sunitinib in wild-type cases is also clear. We have not observed any response to sunitinib in group of patients with PDGFRA mutations (mainly D842V), which has been also shown in preclinical data. We did not analyze the impact of secondary mutations, although patients from clinical trials with tumors harboring a secondary mutation in exon 13 or exon 14 KIT have a longer PFS than patients with exon 17 or 18 mutations [23,[25][26][27].
On the other hand, utility of analysis of secondary mutations is very challenging because imatinib-resistant GISTs are very heterogeneous with multiple clones having different secondary mutations within the same or different nodules [28][29][30]. Sunitinib therapy is associated with several adverse events, which were generally mild to moderate and could be managed by dose modulation (including continuous administration of lower dose) [20,22,24]. The toxicity profile reported in our study is similar to that observed in clinical trials, with exception of hypothyroidism, which occurred in more than 30% of patients (it has been reported outside clinical trials [31,32]). However, up to one third of cases were classified as more severe toxicity (and two deaths due to tumor hemorrhage were classified as related to sunitinib therapy). Our own experience with patients with unresectable or metastatic GISTs, treated with tyrosine kinase inhibitors, suggested the higher incidence of emergency operations for gastrointestinal bleeding, bowel obstruction, or abscess, occur during secondline therapy with sunitinib than during first-line therapy with imatinib [33,34]. This increased incidence of complications leading to surgical interventions with sunitinib could be associated with the presence of more advanced and drug-resistant disease, or to the direct mechanism of action of sunitinib, i.e., the combination of cytotoxic and antiangiogenic activity, leading to dramatic tumor response.
Arterial hypertension is one of the most common complications of sunitinib therapy, occurring usually early after treatment initiation. Serial monitoring of blood pressure is recommended during therapy with sunitinib.  Sunitinib-induced arterial hypertension may also serve as biomarker of antitumor efficacy (probably by antiangiogenic mechanism), because it was an independent factor influencing patient both progression-free and overall survival. Antiagiogenic activity may play an important role in therapy of sarcomas, what has been recently confirmed by positive results of phase III trial with pazopanib in pre-treated soft tissue sarcoma patients [35]. Similar relationships between arterial hypertension induced by VEGF inhibitors (including sunitinib) and oncological outcomes have been reported in renal cell carcinoma patients [15,16,[36][37][38][39]. Treatment-induced persistent hypertension was associated with frequent tumor response, a long time to disease progression and longer overall survival [39]. Clinical outcomes are not compromised by treatment with anti-hypertension medications, moreover, patients who required at least three antihypertensive drugs had the longest PFS and OS [38]. There are proposed some hypothetical mechanisms leading to hypertension related to sunitinib, e.g. presence of less-perfused microvessels and/or diminished number of microvessels, decreasing nitric oxide production and activation of the endothelin-1 pathway leading to vasoconstriction [40,41].
In the subgroup of patients we have analyzed some possible pharmacogenetical relationships with sunitinib tolerance. It has been shown that single nucleotide polymorphisms of VEGF and VEGFR2 genes has some potential as biomarkers for clinical outcomes and toxicity of VEGF pathway targeted therapy [42][43][44][45][46]. We have not studied correlation between SNPs of VEGFA/ VEGFR genes and outcomes of therapy due to limited number of cases, but we have found clear associations between two SNPs of VEGFA gene and sunitinib-induced hypothyroidism. The molecular mechanisms of hypothyroidism induced by sunitinib are unknown, but recent studies have suggested that VEGFR inhibition can induce vasculature regression in various organs, predominantly in thyroid, what can be linked to different properties of VEGF protein caused by gene polymorphisms and sunitinib sensitivity [47,48].

Conclusions
To summarize, we confirmed that many advanced GIST patients benefit from sunitinib therapy with overall survival exceeding 1.5 year. Exploring the toxicity of multikinase targeting agents in GISTs may allow better adjusted therapy as well as to define novel pharmacodynamics markers. Primary tumor genotype and sunitinibinduced arterial hypertension (as surrogate of its antiangiogenic activity) are two independent factors influencing the progression-free survival and OS. The mechanism of side effects and its correlation with pharmacogenetic data during sunitinib therapy need further studies.