2012 ASCO GIST Presentations



Becky, Dr. George Demetri, and Nancy

Becky Bensenhaver and Nancy Berezin once again represented GIST Support International at the huge ASCO meeting to report on the papers and posters presented there for GIST patients. They also roamed the vast exhibit hall, speaking with pharmacutical firms about drugs under development for GIST. On an informal basis Nancy and Becky were able to discuss the future of GIST treatment with experts from around the world. as well as GIST patient group representatives from other countries.

Many thanks to our dynamic duo for volunteering their time and energy to attend the huge meeting and write summaries for us all!

Summaries of many papers and posters relevant to GIST are shown below, each with a link to its abstract on the ASCO website.

Efficacy of a phosphoinositol 3 kinase PI3K inhibitor in GIST models.

Abstract No:10030

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10030)

Authors: Thomas Van Looy, Agnieszka Wozniak, Raf Sciot, Giuseppe Floris, Paul W. Manley, Maria Debiec-Rychter, Patrick Schöffski

This poster described the efficacy of the PI3K inhibitor NVP-BEZ235, alone or in combination with Gleevec (IM) in GIST tumors grafted into mice. Either exon 9 or exon 11 tumors were implanted. The mice were randomized into four groups (control, BEZ only, BEZ plus IM, or IM only). Treatment efficacy was assessed by tumor volume, histopathology and Western blotting. Also, tumor regrowth was evaluated for 3 weeks after treatment cessation.

Results: Apoptotis (cell death) increased significantly after treatment with IM alone (5.5 fold as compared to control group) and IM+BEZ (14 fold) in exon 11 GIST, where it was almost unaffected by BEZ as a single agent, as well as in all treatment groups for Exon 9. By Western Blotting, complete PI3K signaling inhibition was observed only after combination treatment. After treatment cessation, long-lasting growth-inhibition was observed in IM+BEZ treated exon 11 GIST. Moreover, in exon 11 GIST the mitotic index after BEZ alone as well as with BEZ+IM was lower than in the control group even after treatment withdrawal. Conclusions: BEZ shows significant efficacy in GIST xenografts (GIST tumors implanted in mice). Furthermore, combination with IM shows synergistic and long-lasting effects even after treatment withdrawal, which is not the case with drugs routinely used for GIST treatment.

Pharmacokinetics of escalated dose of imatinib in patients with advanced gastrointestinal stromal tumors.

Abstract No: 10085

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10085)

Author(s): Changhoon Yoo, Min-Hee Ryu, Baek-Yeol Ryoo, Inkeun Park, Mo Youl Beck, Yoon-Koo Kang

Escalation of IM dose from 400 to 800 mg/day is a common practice in the management of resistant GIST. However, there is little orno pharmacokinetic (PK) evidence that escalation improves clinical outcomes. This poster explored the relationship between IMplasma trough levels (Cmin), and clinical outcomes and toxicities in patients treated with higher IM dosages. Between 2008 and 2011, Yoo and colleagues measured steady-state IM trough levels in 66 GIST patients receiving escalated IM dosages using liquid chromatography-tandem mass spectrometry. The percent change in Cmin from the standard to the escalated dose was calculated in 43 patients. Association with efficacy and toxicity was analyzed by grouping patients into quartiles according to (1) IM Cmin at the higher dosage level and (2) percent change following dose escalation. Median patient age was 59 years (range, 36–77 years), and the majority of patients were male. KIT exon 11 and 9 mutations were detected in 32 and 18 patients, respectively. Median PFS was 4.2 months and OS was 38.5 months.

Following dose escalation, drug trough levels were significantly increased, with a mean percent change of 162%. Body surface area, hemoglobin level, and neutrophil count were significant variates of IM Cmin. Neither the patient’s Cmin quartile following dose escalation nor the percent change from standard IM dosing was associated with clinical response and survival outcomes. However, there was a significant association between toxicity and Cmin: grade 3-4 hematologic or grade 2-4 non-hematologic toxicity was observed in only 9% (1/11) of patients in the lowest quartile vs 56% (18/32) in the second, third, and fourth quartiles. Interestingly, the clinically significant toxicity that followed dose escalation correlated with percent change in IM Cmin rather than with absolute values. The investigators concluded that (1) raising trough levels of IM did not improve clinical response and survival, and (2)monitoring IM trough levels might help to predict or manage toxicities induced by dose escalation.

Nancy’s comment: This reconfirms earlier observations that more IM is not necessarily better – just more toxic. Becky and I asked Dr. Yoo whether any differences in outcome were seen between patients with Exon 11 and Exon 9 mutations, and he responded that unfortunately the study had not been powered to answer that question.

Randomized phase III trial of regorafenib in patients (pts) with metastatic and/or unresectable gastrointestinal stromal tumor (GIST) progressing despite prior treatment with at least imatinib (IM) and sunitinib (SU): GRID trial.

Abstract No: LBA10008

Citation: J Clin Oncol 30, 2012 (suppl; abstr LBA10008)

Author(s): George D. Demetri, Peter Reichardt, Yoon-Koo Kang, Jean-Yves Blay, Heikki Joensuu, Robert G. Maki, Piotr Rutkowski, Peter Hohenberger, Hans Gelderblom, Michael Gordon Leahy, Margaret von Mehren, Patrick Schoffski, Martin E. Blackstein, Axel Le Cesne, Giuseppe Badalamenti, Jian-Ming Xu, Toshirou Nishida, Dirk Laurent, Iris Kuss, Paolo Giovanni Casali, on behalf of GRID Investigators;

Dr. Demetri presented the Phase III results of the randomized, double-blind, placebo-controlled GRID (GIST-Regorafenib in Progressive Disease) trial, which evaluated the efficacy and safety of Bayer’s new multikinase inhibitor, regorafenib (REG), in patients with metastatic and/or unresectable GIST whose disease had progressed despite prior treatment with IM and SU. Patients were randomized 2:1 to receive best supportive care plus either REG 160 mg once daily (3 wks on/1 wk off) or placebo (PL). The primary endpoint was PFS. Secondary endpoints included OS, disease control rate (DCR, defined as rate of partial response plus stable disease lasting for ≥12 wks), response rate and duration, safety and correlative genotypic analyses. Patients who progressed during the trial period were eligible for crossover to open-label REG. A total of 234 patients were screened and199 were randomized (REG: 133, PL: 66). Median PFS was 4.8 months for REG vs 0.9 months for PL – satisfying the primary endpoint. The most common >grade 3 toxicities in the REG arm were hypertension (28%), hand-foot skin reaction (21%), and diarrhea (8%). The investigators concluded that REG had significantly improved both PFS and DCR, with toxicities as expected for this drug class.

Nancy’s comment: Regorafenib targets C-Kit, VEGFR. PDGFR, RET, RAF-1, BRAF, p38 MAP, and a bunch of other kinases – thus, audience expectations were extremely high. Those hoping to come away from the presentation with news of a dramatic breakthrough were disappointed…but these are early days. Unquestionably, some patients will do very well on this drug. Whether the median PFS can be further improved — and OS increased — remain to be seen.

Single-dimension CT measurements with RECIST 1.1 to evaluate liver metastases in GIST patients on imatinib.

Abstract No: 10091

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10091)

Author(s): Gaia Schiavon, Alessandro Ruggiero, Stefan Sleijfer, Dave J. Bekers, Bronno van der Holt, Jacqueline S.L. Kloth, Gabriel P. Krestin, Patrick Schoffski, Jaap Verweij, Ron H.J. Mathijssen;

Background: Measuring tumor response to therapy is crucial in oncology, both in daily practice and clinical research. RECIST (Response Evaluation Criteria In Solid Tumors) is a set of published guidelines that define when cancer patients improve in response to therapy, remain stable, or progress. The original criteria were published in February 2000 by an international collaboration including the European Organization for Research and Treatment of Cancer (EORTC), National Cancer Institute (NCI) of the United States and the National Cancer Institute of Canada Clinical Trials Group. RECIST 1.1, published in January 2009, is an update to the original criteria. There are four broad response categories:

CR (complete response) = disappearance of all target lesions
PR (partial response) = 30% decrease in the sum of the longest diameter of target lesions
PD (progressive disease) = 20% increase in the sum of the longest diameter of target lesions
SD (stable disease) = small changes that do not meet above criteria

While RECIST represents a major improvement over the confusion that preceded it, it still attempts to describe a three-dimensional, non-uniform object using unidimensional assessments. GIST tumors are shape-shifters, changing their morphology in response to TKI therapy. As techniques for evaluating tumors improve, investigators have begun to question reliance on RECIST, suggesting that more accurate measures, such as computerized assessment of tumor volume and/or functional assessment with PET or MRI, be substituted.

Dr. Schiavon and her colleagues analyzed the morphology of GIST liver metastases (LM) to determine whether unidimensional tumor measurements (per RECIST 1.1) provide an accurate reflection of changes in volume in response to therapy, and to estimate potential bias introduced by using only the longest diameter to measure volume change. Seventy-eight GIST patients with LM were evaluated at baseline (n=139 lesions) and 3 (n=129), 6 (n=128), and 12 months (n=108) after introduction of IM therapy. LM were segmented semi-automatically on standard CT scans. Baseline measurements were obtained by two independent investigators, who were in close agreement regarding segmented volume (mean variability = 1.5%). Calculated volume was also generated using the maximum transaxial diameter (MTD), based on the assumption that metastases are fully spherical. Segmented volume, MTD, and elongation values were reported for each lesion at several time points. At baseline, 44% (61/139) of the LM were spherical and 56% (78/139) were ellipsoidal. During treatment, 42% of lesions kept their morphology and the other half changed from spherical to ellipsoidal or vice versa. The difference between segmented and calculated volume was highly significant (P<0.0001), with calculated volume overestimating actual volume by 35%. The investigators concluded that (1) a lesion’s single largest trans-axial dimension (as used in RECIST 1.1) is not an accurate surrogate of its actual volume, and (2) further studies are needed to determine what type of additional criteria should be applied when evaluating response to therapy.

Nancy’s comment: After reading this poster and chatting with Dr. Schiavon, she had me convinced: The switch from single-dimension measurement to measuring volumetric response with semiautomatic software provides a much more accurate picture of the response of liver mets to TKI therapy than the current RECIST criteria. She further noted that the added costs of using the newertechniques for tumor evaluation are modest, and all the necessary information can be retrieved after the fact from an ordinary CT scan.

Effect of five years of imatinib on cure for patients with advanced GIST: Updated survival results from the prospective randomized phase III BFR14 trial.

Abstract No: 10095

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10095)

Author(s): François Bertucci, Isabelle Laure Ray-Coquard, Binh Bui Nguyen, Antoine Adenis, Maria Rios, Florence Duffaud, Didier Cupissol, Christine Chevreau, Emmanuelle Bompas, Julien Domont, Sylvie Chabaud, David Pérol, Jean-Yves Blay, Axel Le Cesne;

Last year, the BFR14 investigators demonstrated that that interruption of imatinib (IM) therapy in patients with advanced GIST after 1, 3, and 5 years was associated with rapid progression; however, nearly all patients regained tumor control following reintroduction of IM at the 400 mg/day dosage. The 2012 results are more disheartening: Whereas patients in the IM continuation arm of BFR14 have done extremely well, all but one of the patients with residual masses randomized to the interruption arm have relapsed.

Nancy’s comment: GIST experts in the US have tended to downplay the risk of brief treatment interruptions, pointing to the BRF14 data as evidence that nearly all patients who responded to IM initially remain responsive when the drug is reintroduced. However, Dr. Le Cesne – who was sitting in for Dr Bertucci — said that the response tends to be weaker the second time around. The French group feels strongly that any treatment interruption is dangerous in a patient with progressive disease, as it opens the door to rapid tumor growth and increased secondary resistance. Even after six years of IM therapy, he said, “we found no diminution in secondary resistance,” reinforcing the point that IM is a holding pattern rather than a cure for GIST. Subsequently, I found this article in the Annals of Oncology, which states the BFR14 investigators’ position clearly:

http://annonc.oxfordjournals.org/content/early/2012/02/21/annonc.mdr622.full

“In patients with nonprogressive GIST, interruption of imatinib after 1, 3, or 5 years of treatment is associated with a high risk of progression, irrespective of the pattern of initial radiological response or the pattern of radiological response at the time of randomization. These results suggest that imatinib can effectively control tumor progression but may not eliminate tumor cells in patients with advanced GIST. Even patients in complete remission (CR) may have residual tumor, albeit invisible by standard morphological criteria. Therefore, it is not recommended to interrupt imatinib in responding patients, including those with a CR.”

Explored prognostic factors for survival in patients with advanced GIST treated with standard dose imatinib (IM): Results from theBFR14 phase III trial of the French Sarcoma Group.

Abstract No:10092

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10092)

Author(s): David Pérol, Isabelle Ray-Coquard, Binh Bui Nguyen, Antoine Adenis, Maria Rios, François Bertucci, Florence Duffaud, Didier Cupissol, Christine Chevreau, Emmanuelle Bompas, Julien Domont, Sylvie Chabaud, Jean-Yves Blay, Axel Le Cesne;

Prognostic factors for progression-free survival (PFS) and overall survival (OS) were reinvestigated in the entire BFR14 cohort, consisting of 434 randomized and unrandomized GIST patients treated with standard dose IM. Median follow-up was 54 months. As of January 2012, there were 285 progressions (65%) and 161 deaths (37%). Median PFS was 29 months. Four independent prognostic factors were associated with longer PFS:
low tumor volume at inclusion,
performance status = 0,
female sex, and
CD34 positivity on GIST cells.

Five-year OS was 54%. Independent predictors of longer OS included: female sex, performance status = 0, platelet count of <400 giga/L, and CD34 expression on GIST cells. Median PFS and OS in this series were slightly superior to those in the B2222 trial, which the French investigators attributed to the good outcomes of patients who began the study with low tumor volume. They recommended further exploration of the biological significance of CD34 expression on GIST cells.

Early assessment of MCV to predict clinical outcome in patients with advanced GIST.

Abstract No:10086

Citation:J Clin Oncol 30, 2012 (suppl; abstr 10086)

Author(s):Anastasia Constantinidou, Dimitrios Krikelis, David Olmos, Michelle R. Scurr, Robin Lewis Jones, Sue Ashley, Ian Robert Judson;

In patients with advanced GIST, Gleevec (IM) trough levels at a steady state have been associated with clinical benefit. Anecdotal observations have suggested that commencement of treatment with IM may be associated with an increase in red blood cell size as measure by MCV and MCH, indicating the effect of IM on KIT signaling in the bone marrow. This study retrospectively examined a possible connection between changes in MCV (mean corpuscular volume) and MCH (mean corpuscular haemoglobin) after the first 3 months of treatment with IM and PFS. Data for 130 patients with inoperable metastatic GIST treated between 2000 and 2011 were analyzed. Results: In the patients with a 10% or over increase in MCV at the 3 month time point had a significantly longer Progression-free Survival (PFS) as compared to those with a smaller magnitude of change. A similar (although not statistically significant) trend was observed in patients with MCH of 15% or over. Conclusion: Simple laboratory parameters (MCV and MCH are typically present on the routine lab work that GIST patients have performed) may be indicators of IM pharmacokinetics and/or dynamics and may therefore be used as surrogate markers of clinical outcome.

Becky’s comments: This study did not break down the data in regard to different exon mutations, etc. Personally, I noted that my MCV was consistently elevated during my treatment time on IM, even though I experienced primary resistance–but perhaps MCV could be a useful surrogate marker for KIT mutated and PDGFRA mutated GIST.

Masitinib in comparison to imatinib as first-line therapy of patients with advanced gastrointestinal stromal tumor (GIST): A randomized phase III trial.

Abstract No: TPS10102

Citation: J Clin Oncol 30, 2012 (suppl; abstr TPS10102^)

Author(s): Antoine Adenis, Loic Chaigneau, Pierre Michel, Nicolas Isambert, François Bertucci, Binh Bui Nguyen, Olivier Bouche, Laurent Mineur, Tanios S. Bekaii-Saab, Nicolas Penel, Yolene Acin, Alain Moussy, Olivier Hermine

This presentation is an update of an ongoing clinical trial comparing the efficacy and safety of of masitinib (7.5 mg.kg.day) to the control group of imatinib (400 or 600 mg/daily) as first-line therapy (previously untreated patients). Primary endpoint was Progression-free Survival (PFS) with secondary endpoints including Overall-Survival (OS), Time to Progression (TTP) and clinical response rates (RECIST). Patient eligibility criteria include: histologically proven, metastatic or locally advanced non-resectable, or recurrent post-surgery GIST; TKI-naive patient or patient previously receiving imatinib only as a neoadjuvant therapy; and confirmed KIT-positive or PDGFRA-positive tumors. Recruitment is ongoing. Results so far indicate that after 65 months median PFS has not been reached. Percent of patients with PFS at the following number of months were: 12 months – 97%, 24 months – 90%, 36 months- 87%, 48 months – 76%, and 60 months – 62%. Each of these is a higher percent than the corresponding results from the meta-analysis data from imatinib Phase III trials. For a 2011 abstract about this same study see Blay, JCO 29: suppl 4 abstract 85.

Masitinib in imatinib-naive advanced gastrointestinal stromal tumor (GIST): Five-year follow-up of the French Sarcoma Group phase II trial.

Abstract No: 10089

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10089)

Background: Masitinib is a tyrosine kinase inhibitor that, in vitro, has greater activity and selectivity than Gleevec (IM) against wild-type c-kit receptor, against activated mutant c-Kit found in GIST such as exon 11 and exon 9, and against resistant mutants (secondary resistance) such as exon 14. Masitinib does not inhibit kinases that are linked to toxic events such as Abl and Src. Objectives: IM-naive patients with inoperaple advanced GIST received oral masitinib until progression without clinical benefit, refusal or toxicity. Efficacy variables included response rate, best response (RECIST), Progression-Free Survival (PFS), and Overall Survival (OS). Presented here are 5-year follow-up safety and survival data. 16 patients were included in the study. The genomic analysis of the patients was: Wild-type 19%, Exon 11 56%, Exon 11 + Exon 13 13%, PDGFRa/Other 6%/6%.

The median treatment exposure was 33.5 months (cut off date: May 2, 2012). 5 patients (20%) were still under study treatment at the time of the cut-off date. Of those, 4 were ongoing without progression or tumor resection. 1 patient was ongoing with tumor resection. 25 patients (80%) had withdrawn from the study. 15 patients with progression (RECIST). 5 patients with adverse event. 2 patients were wrongly included (ueterine stromal sarcoma and fibromatosis). 1 patient with appearance of prostatic cancer. 1 patient with liver mets, became resectable. 1 patient investigator decision.

Five patients showed prolonged response: 4/5 had KIT exon 11 mutation (1 missing data), 3/5 experienced complete response and 2/5 partial response.

Single agent masitinib at 7.5 mg/kg/day was well tolerated: AEs were frequent (30/30 pts), albeit mild most common AEs were: asthenia (87%), diarrhea (57%), eyelid edema (57%), nausea (47%), muscle spasms (43%), and rash (37%). 14 pts (47%) reported at least one grade 3 or 4 related AE. The most common treatment related grade 3 toxicities were: rash (10%), and neutropenia (7%), with on patient presenting a grade 4 skin exfoliation (3%). No related deaths. No related life-threatenining events.

Historical comparison with imatinib: Despite the limited validity of comparison with phase II trials, OS and PFS with masitinib appear to compare favorably to those of imatinib at 400 mg.:
Median PFS (months): Masitinib: 41 Imatinib: 18
Median OS (months): Masitinib: 65 Imatinib: 55

Historical comparison of OS and PFS (exon 11) from the masitinib phase II study with the imatinib phase III study:
Median PFS (months): Masitinib: 45 Imatinib: 27
Median OS (months): Masitinib:Not reached Imatinib: 60

Conclusions: This 5-year follow-up data substantiates that masitinib has an effective and sustainable activity in IM-naive GIST patients. Despite the limited validity of comparison with phase III trials, the median PFS and OS of masitinib compare favorably to those of imatinib. Adverse events occurred mainly during the first year, particularly over the initial 3 months, with good long term tolerance experienced thereafter. These results support the head to head comparison with imatinib in the ongoing phase III randomized clinical trial in first line advanced GIST patients.

Immunohistochemical identification of SDHA-mutant gastrointestinal stromal tumors (GISTs).

Abstract No: 10029

Citation: J Clin Oncol 30, 2012 (suppl; abstr 10029)

Author(s): Andrew J. Wagner, Stephen P. Remillard, Yixiang Zhang, Jason L. Hornick

Discussion: A subset of WT GISTs have loss of expression SDHB. Only a fraction of SDHB-defiecient GISTs carry loss-of-function mutation in SDHB or SDHC. Due to the complexity of its locus and the presence of several pseudogenes, SDHA is rarely analyzed. Recently, mutations in SDHA were shown to lead to loss of expression of SDHA and SDHB in paragangliomas. This study sought to determine whether SDHA IHC could identify GISTs with SDHA mutations. Results: SDHA mutations are a common cause of SDH-deficient GISTs and result in combined loss of expression of both SDHA and SDHB. Loss of SDHA expression by IHC reliably predicts SDHA mutations and can be used to select cases for SDHA genetic testing.

Becky’s comment: I found this information to be interesting. Dr. Kim had indicated as he ended the last NIH Clinic that they were in the process of analyzing all of the present and (hopefully) previous clinic participants samples for SDHA mutations. The study above indicates that SDHA IHC staining could be a reliable indicator of the presence of SDHA mutations.

——————————————————————————–