What dose Gleevec is best?

It isn’t as simple as measuring the level of Gleevec in your system. A host of factors — from size to age to stress — may come into play

By Jerry Call

The debate over what dose of Gleevec will best prevent disease progression continues, as the initial results of the two large phase III trials continue to show crucial differences. The Europe/Asia/Australia study is finding a longer time to progression at 800 mg. vs. 400 mg. The U.S./Canada study has not found any such correlation between dosage level and time to progression. A recent Life Raft Group analysis of patient-reported data showed fewer relapses at higher doses -- especially when the analysis was based upon actual dose delivered instead of the starting or “intent to treat” dose.

In a recent presentation given by Dr. Allan Van Oosterom at the Life Raft Group meeting in Orlando, Fla., the Life Raft Group learned that the Europeans have been monitoring blood levels of Gleevec in phase III patients and that these levels tend to decrease over time. It has been speculated that this may be why patient side effects often lessen over time. The study differences, as well as the declining Gleevec levels in the blood, has led some patients to ask whether all patients’ blood levels should be monitored. This article intends to provide a foundation to more fully understand this question.

Two important areas in the study of drugs are pharmacodynamics and pharmacokinetics. Pharmacodynamics (PD) is basically “what the drug does to the body.” Pharmacokinetics (PK) is “what the body does to the drug.” The introduction of a few important PK concepts may be helpful to the reader.

CLEARANCE

Clearance (CL) is one of the primary PK parameters. Clearance describes the efficiency of elimination of a drug from the body. It is defined as “the volume of blood cleared of drug per unit time”. As an example; if liver blood flow is 90 liters (L) per hour and clearance were 60 L/hour, then about two-thirds (60/90) of the drug entering the liver is removed or cleared by the liver in one pass. A higher clearance rate means the drug is being removed from the body faster. Variations in clearance from one person to another or in the same person over time might require an adjustment in dose. Drug interactions can also affect clearance. This is the basis of the well-known “drugs to avoid” list that was distributed in the early days of the phase II trials. Gleevec is metabolized in the liver by at least two enzymes, CYP3A4 (primary) and CYP2D6 (secondary). Other drugs and even “natural” supplements can affect these enzymes, causing either an increase in activity (an inducer) or a decrease in activity (an inhibitor).

Drugs that induce either CYP3A4 or CYP2D6 can cause a decrease in Gleevec drug concentrations by increasing clearance of Gleevec from the body. This could result in not enough Gleevec to prevent tumor progression. This type reaction might be more worrisome for a patient on a lower dose of Gleevec. It might also concern a patient who was experiencing a decrease in side effects.

Drugs that inhibit either CYP3A4 or CYP2D6 can cause an increase in Gleevec drug concentrations by reducing the clearance of Gleevec from the body. This could result in higher levels of Gleevec in the body, possibly resulting in increased toxicity. This might be more of a concern for a patient already on a higher dose of Gleevec or one who was having significant side effects.

Variations in CYP3A4 and CYP2D6 result in patients metabolizing Gleevec at different rates. This means different levels of Gleevec in different patients. Data supplied to the U.S. Food and Drug Administration showed drug concentrations varied up to 40 percent between patients taking the same dose. Some other sources cite higher variation, with one source citing variability up to four-fold. (ASCO 2003, Imatinib Elimination: Characterization by in vivo testing of phenotype and genotype, http://www.asco.org/ ac/1,1003,_12-002511- 00_18-0023-00_19- 002003,00.asp). Polymorphisms in the ABCB1 gene (MDR1) appeared to be the primary cause of the variation in this study.

Data submitted to the FDA suggests that clearance of Gleevec is related to both age and weight. (However, another study of CML patients found that weight did not seem to be a factor in clearance.) The information from the FDA suggested that older age and lighter weight seemed to reduce clearance (increasing Gleevec blood levels) and that younger patients and heavier patients tended to clear Gleevec faster (decreasing blood levels).

Variations in protein binding to Gleevec in the blood may also affect clearance. More will be said about this later.

DRUG CONCENTRATION

“C” designates the concentration of a drug in the body. Cmax is the maximum concentration (occurring once a day when taking pills a per day). This occurs about two to four hours after taking Gleevec. Cmin is the minimum concentration of drug. The minimum concentration occurs at about the time (or just before) Gleevec is taken.

HALF-LIFE

The half-life (t1/2) of a drug is the time taken for the amount of drug in the body (or the blood) to fall by half. It is a composite PK parameter determined by both clearance (CL) and volume of distribution (V). Although volume of distribution is an important PK parameter, it is a little beyond the scope of this article.

Half-life is important because:
• It determines the duration of a drug’s action after a single dose. The longer the half-life, the longer the drug concentration stays in the effective range.
• It determines the dosing frequency. The fluctuation in drug concentrations is determined by the half-life and the frequency of doses. A drug with a shorter half-life requires more frequent doses.
• It determines the time required to reach steady state drug concentrations with chronic dosing. It typically requires about three to five half-lives to reach steady state concentrations.

PARENT DRUG and METABOLITE

There are at least two important versions of Gleevec in the body. The first is unchanged Gleevec (called the “parent drug”). It normally accounts for about 75 to 85 percent of the total active Gleevec in the body.

The second form of active Gleevec is a metabolite called CGP 74588 (a form of Gleevec altered by the body’s metabolism). CGP 74588 accounts for the remaining Gleevec in the body (about 15 to 25 percent). Both of these Gleevec components have approximately equal anti-tumor effects.

Another difference between these two forms of Gleevec is that they have different half-lives. The parent drug (Gleevec) has a half-life of about 18 hours while the metabolite (CGP 74588) has a half-life of about 40 hours. Both of these have considerable variation in half-lives between different patients. The effect of half-life is more important if you are trying to measure Gleevec levels in the first few weeks of exposure to Gleevec, and become less important to measuring levels as time goes by.

PROTEIN BINDING

Another important, and often overlooked, point is that all drugs bind to other proteins in the blood and only a fraction of the drug is actually available to enter tissues and tumors. The amount of protein binding varies greatly among drugs. It is not the amount of protein binding that causes problems in determining standard dosage levels; it is the variation in protein binding between patients that could cause problems.

Gleevec is approximately 95 percent protein-bound. That means that 95 percent of the Gleevec that a patient takes is stuck (bound) to a protein in the blood, and that a mere 5 percent of the Gleevec is unbound or “free” Gleevec. Only free Gleevec is available to enter tissues/tumors. If everyone had 95 percent of their Gleevec bound to protein, then protein binding would not matter and measuring Gleevec levels in the blood might be adequate by itself to determine if a person was getting adequate Gleevec. The amount of protein binding depends on a number of things including the drug concentration.

In personal communications with Dr. Bin Peng of Novartis, he states that protein binding of the CGP 74588 metabolite of Gleevec is approximately the same as the parent drug.

Dr. Carlo Gambacorti-Passerini has done a number of experiments that show that high levels of alpha-1-acid glycoprotein (AGP) can cause increased Gleevec binding to AGP and thus reduce the amount of “free Gleevec.”

AGP is an acute-phase protein that is increased when the body is under great stress. This especially includes infections, but also includes cancer, inflammation and post surgery. It has been suggested that larger tumor burdens generally result in higher levels of AGP.

The table above is from Gambacorti- Passerini’s paper “Alpha-1-acid Glycoprotein Binds to Imatinib (STI571) and Substantially Alters Its Pharmacokinetics in (CML) Patients.” It shows the effects of inter-patient variability of protein binding. All of these patients were on 400 mg. of Gleevec. These patients were treated with clindamycin for established infections or for prophylaxis. Since they had infections, their levels of AGP might have been higher than normal.

If we look at the second column (Cmax) , we can see that maximum drug levels varied by 3.25 fold. The sixth column shows that most of these patients had higher protein binding than reported as average/ typical (95 percent is typical). The last column shows that free Gleevec levels varied by 5.2 fold.

The role of AGP in the pharmacokinetics of Gleevec remains controversial and computer analysis suggests that AGP levels do not affect the amount of free Gleevec available (personal communications with Novartis).

A commercial test to measure AGP levels is available. One Life Raft Group patient has been monitoring AGP levels and they have increased over time as follows:

As this patient’s AGP levels have increased, so has the amount of drug that the patient has been able to tolerate. This patient started at 600 mg., was reduced to 400, then 300 mg. for side effects, and today is able to tolerate 700 mg.

Gambacorti-Passerini also found that Gleevec drug concentrations tended to be higher in CML patients that had higher AGP levels. This finding seems to find support in a 2004 ASCO abstract “Inflammatory response affects the pharmacokinetics (PK) and pharmacodynamics (PD) of imatinib and CGP 74588 in patients with advanced gastrointestinal- sarcoma (GIST)”, by C. Delbaldo and others. Although the conclusions of this abstract are difficult to interpret, they seem to have found that high AGP levels decrease clearance of Gleevec and CGP 74588 (and therefore would increase the total levels of Gleevec). The higher drug levels in this group may however, be misleading. Along with the higher AGP levels, protein binding may increase offsetting the higher drug levels. For that matter, protein binding may more than offset the higher drug levels, possibly resulting in ineffective amounts of “free Gleevec” even though total blood levels would seem to be adequate. What is uncertain from these two studies is to what degree the higher drug levels offset the higher protein binding.

Three main proteins, Erythrocytes (red blood cells), albumin, and AGP, appear to be largely responsible for variations in protein binding between patients. Dr. Ian Judson and the European groups have reported that hemoglobin (a protein related to red blood cells) and albumin both affect response/side effects. They report that this appears to be due to their known influence on Gleevec pharmacokinetics. They do not report on AGP and perhaps, they did not include AGP in their analysis.

There may be a potentially important difference in protein binding between AGP and either erythrocytes/hemoglobin or albumin. The difference is in the direction of variation from normal. Patients that have variations from normal in erythrocytes/hemoglobin and albumin tend to have lower than normal values.

Theoretically, a patient that is varying from normal might actually have less protein binding than normal, potentially having more drug/more side effects than normal. On the other hand, AGP tends to vary from normal to levels that are up to four to five times higher than normal in some cases. So a variation in AGP from normal would tend to increase protein binding, potentially causing a reduction in free drug AND a reduction in the efficacy of the drug.

One thing to be aware of that is very important is that in the early CML trials, a better correlation was found between dose and response than either Cmax or Cmin and response. So this leaves us with the question: Should we be monitoring blood levels? Given care to interpret the results, monitoring levels might be useful in the following places:
• To follow a trend in blood levels. Fox example, if clearance decreases over time as reported by the Europeans.
• To help detect drug interactions. We have seen from the early phase I combination trials that Gleevec may interact with other drugs, as it does with PKC412 where Gleevec levels are dramatically reduced. Many, if not most, GIST patients also take other drugs.
• They might be useful as a general tool to tell if a patient were getting adequate amounts of drug IF a method to estimate or measure the effects of protein binding is developed.

Disclaimer: This report was drafted by Jerry Call, science coordinator of the Life Raft Group, and edited by Norman Scherzer, executive director of the group. Neither are physicians, scientists or professional researchers. The article is intended to provide basic information about a very complex but important subject area, and to provoke thought and comment. As survival time for GIST patients is of the essence, our intent is to push the envelope of scientific thought so as to hasten the day when GIST will be considered a chronic disease. We welcome corrections and comments.