Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

MEDICAL FRONTIERS - 2008 Gastrointestinal Cancers Symposium

Orlando, Florida / January 25-27, 2008

Prior to the era of the tyrosine kinase inhibitors (TKIs), slightly over half of patients with primary gastrointestinal stromal tumours (GIST) without metastatic disease would still be alive five years after surgery. This indicated that despite a successful complete resection, prognosis remained poor. That prognosis was worse in cases of patients with tumours >10 cm or if the tumour was located in the small bowel. However, by far the most dominant predictor of survival in patients with GIST is mitotic rate, the difference in five-year survival being considerably higher at about 80% when the mitotic rate is low to only about 20% when the mitotic rate is high. Investigators are also paying more attention to mutational status as another potentially important marker of individual risk.

As discussed by Dr. George Demetri, Associate Professor of Medicine, Harvard Medical School, Boston, Massachusetts, one of several major advances made in the realm of GIST over the past few years is the recognition that GIST is not just one disease, but rather several distinct molecular subtypes.

“We know that GIST represents a wide array of histologic, molecular, ultrastructural and immunophenotypic manifestations,” he told the audience. It is also now widely recognized that GIST is related to the receptor tyrosine kinase kit and its signalling by functional biology. Kit-expressing GISTs also exhibit a wide variety of mutations—in exon 9, exon 11, exon 13 and even “wild-type” tumours, where there is no activating mutation but, despite the absence of such mutations, “kit is unusually, uncharacteristically and uncontrollably turned on and is signalling,” as Dr. Demetri observed.

Unlike other cancers such as those of the colon or lung, a single dominant mutation is always present in GIST patients. However, as Dr. Demetri also noted, uncontrolled kit signalling is not the sole element in the development of GIST and there is likely a “secondary hit” that provokes GIST growth which is not yet well understood. For example, it has long been held that GISTs have a high level of malignant potential, even though a tumour of <1 cm in size is highly unlikely to recur or metastasize. Rare patients with familial GIST are also born with kit mutations but they do not necessarily present with large tumours until they are elderly.

Several investigators have also found evidence of micro-GISTs during autopsies of stomach cancer patients but whether micro-GISTS have malignant potential is not known. What is clear is that clinical progression in patients with GIST is accompanied by serial, timed acquisition of cytogenetic changes, Dr. Demetri explained. Following the discovery of c-kit mutations as the key oncogenic signal in the development of GIST, Dr. Demetri reminded delegates about the remarkable survival advantage now achievable in the era of TKIs, with imatinib as the first of its class to be approved for metastatic GIST.

Studies consistently indicate that in the metastatic setting, over 80% of patients derive clinical benefit, essentially tripling survival over previous treatment standards. In a recent analysis of the benefit of imatinib derived from two randomized studies, Dr. Demetri and European colleagues examined recurrence-free survival (RFS) rates according to both genotypic status and dose (either 400 mg or 800 mg/day).

At a median follow-up of 45 months, the MetaGIST analysis showed that patients with tumours carrying an exon 11 mutation had the best outcomes with a median RFS and median overall survival (OS) of 26 and 60 months, respectively. Patients with tumours that did not contain active mutation fared favourably with a median RFS and OS of 16 and 43 months, respectively, while those whose tumours expressed exon 9 mutations fared the worst with a median RFS of 13 months and a median OS of 31 months.

Interestingly, patients with exon 9 mutations who received the higher dose (800 mg) had a longer RFS interlude than those treated with standard-dose imatinib but OS was not significantly different between the high and lower dose groups and dose did not affect outcomes for patients with any other mutational status.

Even though the kit signal can be completely suppressed with initial treatment, it is usually reactivated over time, at which point patients become resistant to treatment and progress. Other TKIs including sunitinib can sometimes override imatinib-resistant disease, Dr. Demetri observed, but not in tumours with an exon 17 mutation.

“Almost any of the TKIs will show some activity in the setting of resistant disease,” Dr. Demetri indicated, “but we are seeing a remarkable emergence of different resistant clones and this really supports the concept that we are going to have to move to combinations to really help our patients.”

Plasma Levels

In a study presented later on during the meeting by Dr. Demetri, investigators performed a pharmacodynamic analysis of imatinib to correlate it through plasma concentrations with clinical response. The study originally involved 147 patients with unresectable or metastatic GIST who were randomized to imatinib at either 400 or 600 mg/day.

Plasma levels were analyzed in 73 patients and patients were grouped into quartiles according to trough plasma concentrations: Q1, Cmin <1110 ng/mL; Q2 and Q3, Cmin ³1110 to <2040 ng/mL; and Q4, Cmin ³2040 ng/mL.

Of these 73 patients, approximately 29% were still in the study at five years. An overall response was achieved in eight out of 18 patients in the Q1 group (44%) compared with 24 out of 36 patients (67%) in the Q2-Q3 group and 14 out of 19 patients (74%) in the Q4 group. Median time to progression was 11.3 months for patients in the lowest trough quintile vs. 30.6 months for those in the second and third quartile and 33.1 months for those in the highest quartile (Table 1). In contrast, no significant differences in median OS were seen between the quartiles.

Table 1. Plasma Levels Correlate with Clinical Benefit

Investigators concluded that monitoring pharmacokinetic or pharmacodynamic relationships might prove to be novel predictive markers for outcome in GIST patients and that exposure to trough plasma levels of imatinib in excess of 1110 ng/mL is important for optimal clinical response.

Adjuvant Therapy

Given imatinib’s efficacy in metastatic GIST, it was a reasonable hypothesis that adjuvant therapy might decrease the risk of recurrent disease in patients with completely resected primary GIST. Dr. Ronald DeMatteo, Vice Chair, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, presented findings he had originally discussed at ASCO 2007. He reviewed results from the ACOSOG Z9001 adjuvant trial, which proved to be of great interest.

ACOSOG Z9001 was a randomized, double-blind, placebo-controlled trial in patients with a primary GIST tumour of 3 cm or more.

Following complete gross resection, those with kit- positive tumours received imatinib 400 mg (n=325) for one year or placebo (n=319). Between 36% and 40% of patients in both arms had a tumour of 3 to 6 cm; between 34% and 37% had tumours ranging from 6 to10 cm; and approximately one-quarter had tumours >10 cm. On recurrence, treatment assignment was unblinded and patients were allowed to cross over to active treatment if they had been on placebo. If they had been on imatinib, the dose was increased to 800 mg.

As Dr. DeMatteo noted, accrual to the trial was halted on the recommendation of the ACOSOG external data monitoring committee based on results of a planned interim analysis of 644 evaluable patients. At a median follow-up of 13 months in recurrence-free patients, 97% of patients receiving imatinib were free of recurrence compared to 83% of those receiving placebo (unadjusted P=0.0000014).

In the two groups with intermediate and large tumours, RFS was also significantly longer in the active-treatment arm vs. placebo although the difference in RFS rates was not statistically significant in patients with the smallest tumours. Because follow-up was prematurely halted by the Z9001 data safety monitoring committee, OS (the trial’s secondary end point) was not significantly different between active and placebo groups.

“Our median follow-up for this trial was only 13 months but it’s still very impressive that almost nobody died within the first year,” Dr. DeMatteo observed. Toxicities were more frequent in patients receiving imatinib, although grade 3 toxicities were seen in 15% of placebo controls and as well as few grade 4 toxicities. Typically, grade 3 and 4 toxicities were hematologic in nature and in general, the TKI is reasonably well tolerated, Dr. DeMatteo indicated.

At this year’s meeting, Dr. DeMatteo also presented data from ACOSOG Z9000, an open-label, multicentre phase II study involving 107 evaluable patients, all of whom underwent complete resection of primary GIST and whose tumours were kit-positive. Patients were deemed to be at high risk of recurrence due to tumour size (³10 cm), tumour rupture or fewer than five peritoneal metastases. Standard-dose imatinib was again administered for one year.

At a median follow-up of four years, 99% of patients were still alive at one year and 97% were alive at two and three years (Figure 1). RFS rates were 94% at the end of the first year, 73% by the end of the second year, and 61% after three years (As Dr. DeMatteo mentioned, typically only about 20% of patients with a 10-cm tumour are alive at five years, an encouraging result, in comparison to historical controls) (Figure 2). The TKI was again well tolerated, with over 80% of patients completing therapy.
00: Overall Survival

<img1476|center>

Results from both Z9001 and Z9000 were also “basically superimposable,” as Dr. DeMatteo also noted—a “striking” observation, he added, “and this raises the possibility that maybe some of these patien
nt therapy.”

Figure 2. ACOSOG Z9000: Recurrence-free Survival

<img1477|center>

Neoadjuvant Administration

Most speakers did not address the issue of unresectable disease, although neoadjuvant administration of imatinib would appear to have a potentially useful role in helping shrink tumours to the point where they might be more amenable to surgery. In June 2006, a phase II trial was designed to try and better assess the usefulness of the neoadjuvant approach in patients who were felt to be either unresectable or who had tumours in difficult areas. Surgery was then followed with adjuvant imatinib for two years.

At approximately two years’ median follow-up, investigators have yet to report their findings but preliminary results suggest that in those patients whose GISTs can be reduced with imatinib and who subsequently undergo a complete surgical resection, the median RFS at two years is over 80%, as reported by Dr. Burton Eisenberg, Professor of Surgery, Dartmouth Medical School, Hanover, New Hampshire. The TKI is also being studied in EORTC 62024, a phase III study for intermediate or high-risk patients who will receive treatment for two years, and SSGXVIII, a phase III study for high-risk patients who will receive imatinib for either one or three years.

Other Treatments

The novel TKI sunitinib has also been approved for the treatment of metastatic GIST. As reported here by Dr. Jeffery Morgan, Instructor in Medicine, Harvard Medical School, 1091 patients with advanced GIST who had failed on imatinib therapy were involved in an ongoing open-label study that allowed them access to sunitinib given at 50 mg/day, in six week cycles—four weeks on treatment, two weeks off. Patients had a median of four treatment cycles, with a median follow-up of 261 days.

Approximately 40% of patients required a reduction in dose, while 57% of patients required an interruption in their schedule, the majority due to adverse events (AEs). The most common grade 3 and higher AEs were fatigue, abdominal pain and hand-foot syndrome, while hypothyroidism occurred in 7% of patients.

However, as Dr. Morgan noted, the safety profile seen in this study was similar to that seen with sunitinib in a prior phase III GIST study reported by Demetri et al. (Lancet 2006;368:1329-38), where most AEs were mild to moderate in severity.

At data cut-off, 697 (64%) patients were still alive, with a median time to progression of 37 weeks. The median estimated OS was 73 weeks. Median survival in patients who had prior imatinib at 400 mg or less was 93 weeks, while for those who had prior imatinib at greater than 400 mg/day, the median survival was 68 weeks. To date, adjuvant trials with sunitinib in GIST have not been reported.

Questions and Answers

The following section is based on discussions with Dr. Ronald DeMatteo, Vice Chair, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York.

Q: OS was not significantly different between active and placebo groups, primarily because follow-up was prematurely halted as per the recommendation of the Z9001 Data Monitoring Committee. With this in mind, would you consider a longer RFS rate a valid end point in and of itself?

A: It depends on your philosophy about treating patients with cancer. Is it acceptable to delay recurrence or even decrease the risk of recurrence by using the drug continuously, or should we wait for patients to recur, knowing that survival might be the same in the long run? So it depends on what your primary end point is—RFS or OS. I think if I were a patient, I’d want to be recurrence-free as long as I could, even if I had the same chance of survival, but I think it’s an individual choice.

Q: You mentioned that with the encouraging RFS rate in the adjuvant setting, you might potentially be curing patients. Would this argue strongly in favour of adjuvant imatinib?

A: In a nutshell, yes. There is a potential for curing at least some fraction of these patients and even if imatinib doesn’t cure them, it may prolong their time to recurrence. The down side is cost and the fact that we don’t know how long they should be treated or which patients should be treated selectively.