By Neil Sandson, M.D.
[Spring 2004; Vol. 30, No. 3; Pg 5, 14]
In the first installment of my two-part discussion of this topic, I explored drug-drug interactions (DDIs) involving the cytochrome P450 system, which is an example of phase I, or oxidative, metabolism. A related system is phase II metabolism, which involves conjugation. Most of phase II metabolism relates to the family of uridine 5’-diphosphate glucuronosyltransferases (UGTs), which perform glucuronidation and which renders compounds more hydrophilic, and thus more readily excretable. As with the P450 system, there are different enzymes that perform glucuronidation, and these are named using an analogous number-letter-number scheme (1A1, 1A4, 2B7, 2B15, etc.). There are also substrates, inhibitors, and inducers involved in phase II metabolism.
Phase I metabolic systems, like the P450 system, perform most of the work involved in drug metabolism, so these systems will generally be a more relevant factor in DDIs than phase II metabolism. However, there are several agents whose metabolism is handled solely or in large part through phase II glucuronidation, such as lamotrigine, olanzapine, lorazepam, and numerous NSAIDs and narcotic analgesics. It’s a good idea to be familiar with the major inhibitors and inducers of these and related agents to avoid DDIs when using these drugs. One of the most complete phase II glucuronidation tables in existence can be found at www.mhc.com/Cytochromes.
Let’s examine a few case examples illustrating the importance of phase II metabolism:
Case 1: A 25-year-old woman with a long-standing seizure disorder had been seizure-free for 5 years while taking lamotrigine (Lamictal), 200 mg/day. She then started taking a standard oral contraceptive containing ethinylestradiol. Events proceeded uneventfully until she had a grand mal seizure 3 weeks later.
Discussion: This is an example of an inducer added to a substrate. Lamotrigine is primarily metabolized through phase II glucuronidation, specifically by the UGT 1A4 enzyme. Ethinylestradiol is an inducer of UGT 1A4. With the introduction of the ethinylestradiol, more UGT 1A4 was produced and thus available to more efficiently metabolize the lamotrigine. This led to a decrease in the blood level of lamotrigine, causing this patient to experience her first seizure in more than 5 years. One study demonstrated that prescribing oral contraceptives for patients taking lamotrigine led to a mean 50% decrease in lamotrigine blood levels.
Case 2: A 36-year-old woman with bipolar I disorder had been stably maintained on lamotrigine (Lamictal), 200 mg/day (blood level=2.5 micrograms/mL), for the past 18 months. Some persistent dysphoria, anhedonia, and insomnia began to develop, so her psychiatrist decided to introduce sertraline (Zoloft) at a low dosage (25 mg/day). Over the next several days, the patient developed increasing lethargy and confusion. When she reported this to her psychiatrist, he ordered another lamotrigine blood level, which was found to be 5.1 micrograms/mL. After consulting with the affiliated hospital’s pharmacist, the psychiatrist discontinued the sertraline and started the patient on citalopram (Celexa), with no further difficulties of this nature.
Discussion: This is an example of an inhibitor added to a substrate. As mentioned above, lamotrigine is metabolized primarily by the UGT 1A4 enzyme. Sertraline, seemingly unique among the SSRIs, is an inhibitor of UGT 1A4. Thus, the addition of sertraline, even at a low dose, significantly impaired the ability of UGT 1A4 to metabolize the lamotrigine. This led to a sharp increase in the blood level of lamotrigine, doubling it in this case. Since the final lamotrigine blood level was not terribly high (5.1 micrograms/mL), the patient’s resulting symptoms were not too dramatic or distressing. However, it is worth noting that only 25 mg/day of sertraline was able to double the lamotrigine blood level. This would have been more expected if valproate had been the added agent, yet sertraline appears to be every bit as effective as it is in elevating lamotrigine blood levels.
Case 3: A 21-year-old man with Bipolar I disorder was being maintained on olanzapine (Zyprexa), 20 mg/day. He had been compliant with this medication, but during exam week at his college he became acutely manic and required psychiatric hospitalization. He and his psychiatrist agreed on a trial of adjunctive carbamazepine (Tegretol), titrated to a dosage of 1,000 mg/day (blood level=9.6 micrograms/ mL). After about one week on that dose, his manic symptoms had remitted enough to allow a transition to the day hospital. However, 10 days after discharge, he was again grandiose and paranoid. He was rehospitalized. After consulting with the hospital pharmacist, the psychiatrist titrated the olanzapine dosage to 40 mg/day, which eventually produced a remission of psychosis and no side effects.
Discussion: This is an example of an inducer added to a substrate. Olanzapine is primarily a substrate of P450 1A2 and the UGT 1A4 enzyme. Carbamazepine is an inducer of 3A4, 1A2, and UGT 1A4. Thus, over the course of 2–3 weeks, the addition of carbamazepine led to an increase in the amounts of all these enzymes, with the result that the olanzapine was more quickly and efficiently metabolized, causing a corresponding decline in the olanzapine blood level and the reemergence of paranoid and grandiose delusions.
Studies in which carbamazepine was added to olanzapine have produced significant (roughly 40%) decreases in olanzapine blood levels, so the hospital pharmacist advised a doubling of the olanzapine dosage to 40 mg/ day. This compensated for carbamazepine’s inductive effects.
Although the chemical details of phase I and phase II DDIs are different, the patterns that produce these interactions are quite similar. No matter how complex and detailed these cases may seem, they can be simplified into a series of substrate-inhibitor and substrate-inducer interactions. However, this logical cleanliness only applies in these “pure” cases of straightforward alterations in drug metabolism involving two or three agents at a time. When lengthy regimens are compounded by pharmacodynamic interactions and by non-metabolic pharmacokinetic issues like absorption, excretion, and plasma-protein binding, we are faced with daunting complexity. This will become more apparent as we explore the functioning of the P-glycoprotein transporter in the next installment of this series.