March, 2005 – Novartis drug can inhibit secondary KIT mutations

By Jerry Call
Life Raft Group science coordinator

Recently reported research further defines the mechanisms of resistance that GIST tumors use to thwart Gleevec. The researchers also report that the Novartis drug, PKC412, may be able to overcome this resistance in some cases.

Maria Debiec-Rychter was the lead author of the research article in the journal “Gastroenterology.” Peter Marynen, Ph.D., heads up the Center of Human Genetics – Molecular Genetics lab that was involved in this research.

This study investigated the mechanisms of resistance to Gleevec in 26 GIST patients treated at the oncology department of University Hospital in Leuven, Belgium.

The resistance mechanisms noted in this study were similar to those previously reported by Dr. Jonathan Fletcher and others. The KIT protein was activated (phosphorylated) in eight of 10 progressive tumors that could be analyzed during treatment with Gleevec. In half of these cases, KIT was reactivated because the progressive tumors had a second KIT mutation in addition to the original KIT mutation. In the other half, the cause for reactivation of KIT remains unknown.

It was speculated that sequencing of the complete open reading frame of KIT in these samples might identify novel mutations in unexpected regions of KIT. An alternative proposal was that factors influencing drug delivery into the cell or alterations in drug clearance from the body could result in inadequate amounts of drug in the tumor and could cause reactivation of KIT and progression. Thus, resistance could be divided into KIT-dependent (KIT protein was activated) and KITindependent categories, with reactivation of KIT as the most important mechanism of resistance.

The most frequent cause for reactivation of KIT appears to be secondary mutations in KIT. Overall, 12 of 26 patients (46 percent) in this study were found to have secondary mutations in KIT.

PKC41213One patient with a primary KIT mutation (KIT G565R) developed a secondary PDGFRA mutation (PDGFRA D842A) that was not present in the primary tumor. Activation of an alternate receptor has previously been cited as a mechanism of resistance in GIST, but this is the first report that we are aware of that has found a second gene mutation that caused the activation of the second receptor (PDGFRA in this case).



PKC41214Most GIST tumors have mutations in either exon 11 (most common) or exon 9 (second most common). Patients with exon 11 mutations generally have the best response rate to Gleevec. Patients with exon 9 mutations have a lower response rate to Gleevec. When resistance is caused by secondary mutations, patients will generally have at least two types of tumors in their body:
• Those that only have the original mutation (typically in exon 11 or exon 9 of KIT).
These tumors are typically still being controlled by Gleevec.
• Those that have the original mutation in addition to a new (secondary) mutation, typically in exon 13, exon 14, exon 15, or exon 17.

It is these tumors that have two KIT mutations that stop responding to Gleevec.

In some cases, it may just be one or two rogue tumors that can be surgically removed, or controlled a local treatment such as radiofrequency ablation (RFA). Gleevec is then continued to control the remaining, sensitive tumors. In other cases this is not possible, and a clinical trial with a drug active against both types of tumors might be one solution. Another trial to be considered would be combining Gleevec, to control the tumors with only a primary mutation, with a second drug that had activity against the tumors with two KIT mutations.

The secondary mutations found in KIT were; V654A (N=4 tumors), D716N (N=1), T670I (N=3), D820E (N=1), D820Y (N=1), N822K (N=1), and D816G (N=1). The secondary PDGFRA mutation was D842V.

Using a combination of resistant GIST tumor cells, and a Ba/Fe mouse cell line engineered to express mutant forms with KIT primary and secondary mutations, the research team demonstrated that Gleevec was not able to inhibit KIT but PKC412 was able to inhibit KIT in a number of cells containing both primary and secondary mutations.

The Gleevec resistant mutations that were tested were KIT-V654A, KITT670I, and PDGFRA-D842V. All of these proved sensitive to PKC412.These mutations include what may turn out to be the most common secondary mutations (KIT-V654A, and KIT-T670I), and some of the more resistant mutations, KIT-T670I, and PDGFRA-D842V.

The KIT-V654A mutation (in exon 13) was the most common secondary mutation in this small group of patients. In another study at M.D. Anderson Cancer Center in Houston, Texas, U.S., all six resistant secondary mutations noted in the study were this type of mutation.

The KIT-T670I mutation (in exon 14) was the second most common secondary mutation and, in test tube experiments, was inhibited by PKC412. This mutation corresponds to the T315I mutation in bcr/abl in CML patients and the T674I mutation in PDGFRA.

The T315I mutation has proved to be the most difficult mutation to treat in CML patients. For example, the new Bristol-Myers Squibb drug, BMS- 354825, that is proving so effective in Gleevec-resistant CML patients, inhibits basically all known bcr/abl mutants except the T315I mutation.

The other mutation inhibited by PKC412 (but not Gleevec) was the D842V mutation in exon 18 of PDGFRA. While this mutation occurred as a secondary mutation in this series of patients, it is also a primary type of mutation. In the U.S./Finland phase II study, this mutation occurred in two patients (1.6%), or 10% of those that had initial resistance to Gleevec. The D842V mutation is the most common form of PDGFRA mutation. Patients that have PDGFRA mutations tend to have primary tumors located in the stomach.

It is tempting to hypothesize that in this small group of patients with primary D842V mutations, PKC412 given alone might be as effective as PKC412 plus Gleevec, since Gleevec is not effective against this mutation. Thus for patients with initial resistance to Gleevec, clinical testing to identify the type of mutation in KIT or PDGFRA has an approximately 10% chance (this percentage estimate is based on small numbers) or identifying the PDGFRA D842V mutation, which might then suggest PKC412 as a strong clinical candidate drug.

In another study just pre-published online as a Blood First Edition Paper, Joseph Gowney and others reported that PKC412 was able to inhibit a number of other Gleevec-resistant mutations that occur in human mast cell disease and acute myeloid leukemia. The most notable of the mutations inhibited by PKC412 in this study were the KIT D816V and D816Y mutations in the c-KIT activation loop.

Other mechanisms of resistance noted in the Leuven study included:
• Two patients completely lost KIT expression, indicating a KITindependent mechanism of resistance. These tumors no longer stained positive for c-kit (CD 117) and they inverted their histologic appearance from spindle to epithelioid type.
• Two cases where resistance was associated with amplification of KIT or PDGFRA genes. In the PDGFRA case, the patient was resistant to initial Gleevec therapy.

Phase I/II trials combining Gleevec and PKC412 are underway in Berlin, Germany, and Portland, Oregon, U.S. These trials are for GIST patients that are resistant to Gleevec. Lead researchers are Dr. Peter Reichardt at the Robert-Rössle-Klinik, Charité Campus Buch in Berlin, and Dr. Charles Blanke at Oregon Health & Science University in Portland. When given together, a significant interaction between PKC412 and Gleevec has been noted.

Our understanding is that PKC412 causes Gleevec to leave the body faster than when Gleevec is given alone. This combination trial thus requires a higher-than-anticipated dose of Gleevec to achieve normal levels of Gleevec in the body (perhaps 1,000 mg. to 1,200 mg./day). The need to adjust the dose of Gleevec in interaction with PKC412 is one of the major reasons that this trial has moved so slowly.

The Gastroenterology article concluded with the following summary: “… our study confirms the existence of KIT-dependent and -independent mechanisms of imatinib resistance in patients with GISTs and shows novel imatinib-resistant KIT mutant isoforms. It points to the acquisition of imatinib-resistant PDGFRA mutations as a cause of secondary resistance in a KIT-positive tumor and indicates KIT amplification as the possible explanation not only for a secondary but also for a primary resistance to the drug. Moreover, our results prove the sensitivity of KIT-T670I and KIT-V654A and of PDGFRA-D842V mutations to PKC412. Given that individual kinase domain mutations exhibit differential sensitivity to alternative kinase inhibitors, it will be crucial to tailor secondline therapy precisely to the underlying mechanism of resistance.”

Many GIST patients use clinical trials to survive. The challenge is how to match the resistance mechanism that causes Gleevec to fail, to the drug/trial that gives patients the best chance of responding. The current paradigm favors enrolling a diverse group of resistant patients into clinical trials and examining mechanisms of resistance at the end of a trial. Patients often select a trial based on geography, or whatever trial happens to be running at the institution they visit.

There is one advantage in enrolling a diverse patient population into clinical trials; after the trial you can analyze data and you might get lucky and find some patient population that was not expected to benefit, but did. On the other hand, enrolling patients into a trial that has the best rational for benefit also offers some advantages. When a patient does well, both the patient and the trial sponsor benefit.

Note: Dr. Charles Blanke, GIST specialist at OHSU and the principal investigator for PKC412 in the United States, was asked to comment on this article. His response: “The article looks great. You hit the nail on the head. I just gave H/O grand rounds at OHSU and I said that the ultimate goal in GIST is to match the individual patient to his/her best GIST drug …” footnote:
Mechanisms of Resistance to Imatinib Mesylate in Gastrointestinal Stromal Tumors and Activity of the PKC412 Inhibitor Against Imatinib-Resistant Mutants Maria Debiec–Rychter, Jan Cools, Herlinde Dumez, Raf Sciot, Michel Stul, Nicole Mentens, Hilde Vranckx, Bartosz Wasag,,Hans Prenen, Johannes Roesel, Anne Hagemeijer, Allan Van Oosterom, And Peter Marynen (Gatroenterology 2005:128:270-279).