The authors measured plasma fluphenazine levels in 20 schizophrenic patients receiving 25 or 50 mg fluphena- zinc decanoate (FPZ-D) by IM injection every 2 ...
Psychopharmacology
Psychopharmacology (1985) 87:55 58
9 Springer-Verlag 1985
Plasma levels of fluphenazine during fluphenazine decanoate treatment in schizophrenia Sidney S. Chang 2, Javid I. Javaid 1, Maurice W. Dysken 3, Regina C. Casper s, Philip G. Janicak 1'4, and John M. Davis 1'4 1 Illinois State Psychiatric Institute, Chicago, IL 60612, USA 2 Department of Psychiatry, Loyola University, Maywood, IL 60153, USA 3 Department of Psychiatry, University of Minnesota, Minneapolis, MN 55455, USA 4 Department of Psychiatry, University of Illinois, Chicago, IL 60612, USA 5 Department of Psychiatry, University of Chicago, Chicago, IL 60616, USA Abstract. The authors measured plasma fluphenazine levels in 20 schizophrenic patients receiving 25 or 50 mg fluphenazinc decanoate (FPZ-D) by IM injection every 2 weeks. The plasma levels were determined by a sensitive gas-liquid chromatographic (GLC) assay with a nitrogen detector device developed in their laboratory. Using this chemical assay method, they replicated the finding of a sharp initial plasma peak within 24 h after the injection followed by a low but rather stable plasma level as previously reported by nonchemical assay methods. The interval plasma levels (averages of day 4-10 after injection) ranged from 0.17-0.61 ng/ml in 10 patients who received 25 mg; and 0.20-0.93 ng/ml in 7 patients who received 50 mg FPZ-D every 2 weeks. This four-fold variation in plasma levels during FPZ-D injection was smaller than previously reported levels achieved with oral antipsychotic drug treatment. Based on the study of plasma levels achieved with FPZ-D injection and oral FPZ-H (fluphenazine HC1) in 6 patients, the dosage requirement of FPZ-D appeared to be difficult to predict from the oral dosage of FPZ-H in the same patient. Two weeks past injection, fluphenazine was undetectable in approximately half the samples with the GLC method. Thus, radioimmunoassay or radioreceptor assay, which also measures metabolites, might be more suitable for the study of plasma levels in patients receiving FPZ-D injection. Key words: Fluphenazine decanoate - Fluphenazine hydrochloride - Plasma fluphenazine - GLC Initial peak Low plasma levels - Individual variation - Oral and injectable
Fluphenazine decanoate (FPZ-D), a long-acting injectable antipsychotic drug, has been used widely for many years in the treatment of psychotic disorders. However, many questions concerning its dosage requirement and the time course of action remain unanswered. Part of the problem is that fluphenazine (FPZ) plasma levels are extremely low in patients receiving an ordinary dose of FPZ-D injection. However, recent advances in laboratory technique (such as radioimmunoassay and radioreceptor assay), have improved assay sensitivity. Recently our own laboratory has developed a sensitive assay method to measure the plasma level of FPZ using gas-liquid chromatography (GLC) Of)print requests to ." S.S. Chang
equipped with a nitrogen detector device (Dekirmenjian et al. 1978; Javaid et al. 1981). Previously, this assay method has allowed us to detect FPZ in plasma of patients treated with fluphenazine hydrochloride (FPZ-H) (Dysken et al. 1981). We report in this paper the plasma levels of FPZ in patients treated with FPZ-D, and compare these results with reports by others using different assay methods. Several investigators have attempted to examine the magnitude and time course of FPZ plasma levels subsequent to FPZ-D injection using different assay methods. Curry et al. (1978, 1979) and Whelpton and Curry (1976) reported plasma FPZ levels of approximately 0.5 ng/ml by tracing radioactivity in the plasma after injecting 25 mg of C14-1abeled FPZ-D. Using a radioimmunoassay method, Wiles and his associates reported plasma levels of approximately 1.0 ng/ml after a 25-mg FPZ-D injection (Franklin etal. 1978; Wiles and Franklin 1978; Wiles and Gelder 1979). Recently, the use of a radioreceptor assay method has also indicated low plasma FPZ levels (Tune et al. 1980; Harris et al. 1982). In addition to the persistent finding of low plasma FPZ by several investigators using different assay methods, the first two groups (Curry etal. 1978; Wiles and Gelder 1979) also reported a sharp plasma FPZ peak within 24~48 h after FPZ-D injection. In all, these studies provide important information, and since these were nonchemical assay methods, it would be important to replicate these findings using a chemical method such as the GLC method used in our laboratory. Although the intervals between injections may vary, most psychiatrists prescribe FPZ-D injections once every 2 weeks. This practice is based primarily on clinical experience, and it would be useful to measure plasma levels during the 2-week period after injection to see if this approach is supported by a plasma level study. In addition clinicians often give oral FPZ-H to control acute symptoms and then switch to long-acting FPZ-D injection for maintenance therapy. However, the equivalent IM FPZ-D dose for oral FPZ-H is uncertain. It is unknown whether patients need a higher FPZ-D dose if their oral dose of FPZ-H is high. We measured the plasma levels of FPZ in patients who first received oral FPZ-H and then FPZ-D injection in an attempt to address this question more properly. We developed a simple and sensitive acetylation method for FPZ assay by GLC with a nitrogen detection to study plasma FPZ levels over a 2-week period in 20 patients who received various doses of FPZ-D injections. We also studied the plasma FPZ levels in six cases both during FPZ-H and
56 F P Z - D treatments. In summary, we attempted to address the following questions. 1. Can the sharp plasma FPZ peak after IM F P Z - D injection, as previously reported by radioimmunoassay and radioactivity tracing method, be replicated by the G L C method developed in our laboratory? 2. Is the clinical experience that a F P Z - D injedtion lasts for 2 weeks explainable by plasma levels of FPZ ? 3. Is there a correlation between F P Z plasma levels measured during oral F P Z - H and I M F P Z - D treatments in the same patient? Materials and methods
Subjects included 20 schizophrenic inpatients who clinically needed F P Z - D injections. All patients gave written informed consent before entering the study. There were 13 male and 7 female patients whose ages ranged between 18-42 years. All patients were first stabilized with oral antipsychotic drugs including F P Z - H and then switched to F P Z - D injection every 2 weeks. The dosages of F P Z - D given were 12.5 mg in one case, 25 mg in ten cases, 37.5 mg in one case and 50 mg in nine cases. Following the third F P Z - D injection with the same dosage, blood samples of 20 ml were collected from each patient on day 2, 4, 8, 10, and 14 after the injection. Additional blood samples were also collected from 11 patients at 3, 6, and 24 h after the injection to measure the plasma peak. Six patients who were on oral F P Z - H had their plasma FPZ levels measured. The dosage was either 20 mg/day (n = 5) or 40 rag/day (n 1). Blood samples of 20 ml were drawn on day 8, 11, and 14 after the initiation of the fixed oral dose of FPZ-H. The average of these plasma FPZ levels was used as the mean steady-state (S-S) plasma FPZ level. Blood was drawn by syringe and placed in an unstoppered glass tube containing E D T A (ethylenediaminetetraacetate) as anticoagulant. The tube was sealed with Paraffin film before mixing to prevent contact of the sample with the rubber stopper. The plasma was separated by centrifugation and stored at - 16 ~ C until analysis. FPZ was measured by a G L C method developed in our laboratory (Dekirmenjian etal. 1979; and Javaid etal. 1981). Perphenazine was used as an internal standard. After adding perphenazine (20 rig), the 5-ml sample of plasma was extracted with heptane/2-propanol (9 : 1 v: v) by a twostep extraction procedure. The organic layer was dried and acetylating mixture (ethyl acetate/acetic anhydride, 9:1 v: v) was added. The samples containing the acetylated derivatives were heated for 30 min at 65 ~ C with intermittent mixing, then injected (2-5 gl) directly into the gas chromatograph (Hewlett-Packard, model 5730-A). A 0.9-m column (2 m m inner diameter) packed with 2% OV-101 was used. The temperature settings were as follows: column temperature, isothermal 240 ~ C; injection port temperature 300 ~ C; detector temperature 300 ~ C. Helium (30 ml/min) was used as the carrier gas. The method as described could detect 0.2 ng/ml F P Z when 5 ml plasma was used. The coefficient of variation of the method at different concentration was 3.6% for 0.5 ng/ml (n=7), 9.8% for 1.0 ng/ml (n=9), 6.4% for 5.0 ng/ml (n=9), and 7.0% for 10 ng/ml (n--7). Results
Immediately following the F P Z - D injection, there was a rapid increase in plasma F P Z level. Of the 11 cases studied,
Table 1. Plasma levels of fluphenazine (FPZ) after fluphenazine decanoate (FPZ-D) injection
Case
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Dosage (mg)
12.5 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 37.5 50.0 50.0 50.0 50.0 50.0 50.0 50.0 50.0
Plasma FPZ (ng/ml) Initial peak
24~
Average Day 14 (day 4-10)
1.8 4.1 1.8 0.68 0.28 2.42 1.29 0.38 2.2 2.0 0.79 -
0.38 0.70 0.35 0.60 0.28 0.98 0.48 0.11 1.2 0.70 0.57 -
0.38 0.50 0.20 0.34 0.17 0.61 0.42 0.54 0.18 0.57 0.22 0.57 0.43 0.61 0.62 0.26 0.35 0.20 0.93 0.03
0 0.15 0.25 0.36 0.26 0 0.56 0 0 0.20 0.48 0 0.17 0.16 0.52 0
7 cases showed a definite sharp plasma peak within 24 h, 3 cases showed a moderate peak, and I case showed no plasma peak. The highest peak of 4.1 ng/ml was observed in a patient (case 2, Table 1) who received 25 mg FPZ-D. In this case, the plasma peak was 5.9 times (4.1 ng/ml:0.7 ng/ml) the average plasma level observed between days 4-10 after injection. In most cases, the plasma F P Z measured at 24 h was approximately the same level as seen on day 4. The typical time course of plasma level decline in four cases is illustrated in Fig. 1. The plasma level after day 4 remained at low levels and declined at a much slower rate than during the first 24 h. The average plasma F P Z level between days 4-10 was used to represent the interval plasma level between two consecutive injections. The plasma levels before day 4 were excluded because of their proximity to the initial plasma peaks; and the plasma levels after day 10 were excluded because many o f the samples were lower than the sensitivity limit of our assay method. As shown in Table 1, the average plasma FPZ level between days 4-10 ranged from 0.17 to 0.61 ng/ml in 10 patients who received 25 mg F P Z - D ; and from 0.03 to 0.93 ng/ml in 8 patients who received 50 rag. With some exceptions, the average plasma levels of those patients who received 50 mg were higher than those o f the 25-rag group. Excluding an outlier (case 20) whose average plasma level was very close to zero, there was a three- to four-fold difference in plasma levels among those patients receiving the same dosage of F P Z - D injection (Table 1). At the end of 2 weeks, following a F P Z - D injection, the plasma levels were quite low. Table 1 shows that on day 14, regardless of the dosage of F P Z - D given, the plasma levels in 9 of 16 cases were lower than 0.2 ng/ml. A similar trend is seen on day 10, i.e., 8 of 20 patients had plasma levels lower than 0.2 ng/ml. The doses and plasma F P Z levels of 6 patients who first received oral F P Z - H and then IM F P Z - D are listed
57
4.0
2.Oe~ 1.O-
--O
,
T
I 2
i 4
I 6
I 8
I 10
i 12
, 14
25 mg
FPZ-D IM
Time (Days)
Fig. 1. Plasma levels of fluphenazine (FPZ) after 25 mg fluphenazine decanoate (FPZ-D) injection
Table 2. Plasma FPZ during oral FPZ-H and im FPZ-D treatmenff
Case
FPZ-D Injection
Oral FPZ-H
Dosage Plasma FPZ (mg/2 weeks) (ng/ml) Average of day 4-10
Dosage (mg/day)
Steady-state plasma FPZ (ng/ml)
25.0 37.5 50.0 50.0 50.0 50.0
20 20 40 20 20 20
2.1 0.82 1.4 (0.7b) 0.53 0.38 0.35
no.
3 12 13 14 17 I9
0.20 (0.40") 0.57 (0.67a) 0.43 0.61 0.35 0.93
Plasma level converted to 50 rag/2 weeks of FPZ-D b Plasma level converted to 20 rag/day of FPZ-H c Correlation between plasma level during FPZ-D and FPZ-H is not significant (r = - 0.47, P> 0.1)
in Table 2. Since all patients were not on the same dosage the plasma levels were converted to the same dosage bases of 50 rag/2 weeks for FPZ-D and 20 mg/day for F D P - H to facilitate comparison. The plasma levels during oral FPZ-H and IM FPZ-D showed no significant correlation (r= -0.47, P > 0.10). Discussion
After FPZ-D injection, we found that the plasma FPZ levels rose rapidly and peaked within 24 h. Thereafter, they fell to lower levels at a much slower rate. This time course was in agreement with reports by Curry et al. (1978, 1979) using a radioactivity tracing method and by Wiles and Gelder (1979) using a radioimmunoassay method. With more frequent measurements of plasma levels during the first 24 h, Wiles and Gelder (1979) were able to define the peak between 1 8 h after injection. Nasrallah et al. (1978) reported higher plasma levels within 6 h after injection using G L C with an electron capture device. Thus, the results from our G L C method supported the finding of an initial plasma peak after FPZ injection. While most of the reports indicate a similar time course,
the plasma levels differ substantially depending upon the assay method used. In the report of Wiles and Gelder (1979), who used a radioimmunoassay method, the interval mean of plasma FPZ levels (excluding the peak) in 12 patients who received 25 mg FPZ-D, varied from 1.02-3.48 ng/ml. With the same dose, Curry et al. (1978) reported plasma levels around 0.5 ng/ml. Using a radioreceptor assay method, Tune et al. (1980) reported mean serum levels of 0.40.8 ng/ml in absolute FPZ concentration in 2 patients who received 35 mg FPZ-D. Our values of 0.17-0.60 ng/ml were closer to those reported by Curry eta]. (1978). In the reports of Curry et al. (1978, 1979), metabolites such as FPZ sulfoxide and 7-OH-FPZ were separated from FPZ before measuring the radioactivity. In essence, our method and Curry's method were measuring FPZ only, while the radioimmunoassay technique also measured metabolites, which may have contributed to the higher plasma levels they reported. Another factor that can affect plasma FPZ levels is the time when blood samples are drawn. It takes four to five injections at regular intervals to reach "steady-state". In this study the plasma levels were measured after the third of three injections given at 2-week intervals and, therefore, steady-state had not been achieved. It is conceivable that if we measured the levels after a few more injections, the plasma FPZ levels might have been higher than reported in this study. Among those patients who received the same dosage of FPZ-D, we found a three- to four-fold difference in the average plasma levels. In an oral FPZ-H study reported by Dysken et al. (1981), there was almost 40-fold variability in mean S-S plasma levels in 21 patients treated with 20 mg/day FPZ-H. Although the number of subjects receiving the same dose of FPZ-D in this study was small, the interindividual differences in plasma FPZ levels on FPZ-D injection seemed to be much smaller than during oral FPZ-H treatment. Even though all the reports do not agree, Wiles and Gelder (1979) also found a four fold difference in plasma levels after the same dosage of FPZ-D injection, thus supporting our findings. Our failure to find a correlation between plasma FPZ levels during oral FPZ-H and IM FPZ-D is in agreement with the report of Harris et al. (1982). They further stated that this might be due to the individual variation in the relative absorption, distribution, metabolism, or excretion of each of the preparations. Since the number of subjects is small in both of these two studies, this finding should be regarded as preliminary until a larger sample can be collected. Near the end of the 2-week period after injection, the plasma levels were extremely low in many of our cases. This finding is not consistent with the clinical experience that FPZ-D injections can usually be given at an interval of every 2 weeks. In many instances, the interval can be extended to 3 weeks or longer with satisfactory therapeutic effect. Our GLC method only measured the parent compound. The radioimmunoassay and radioreceptor assay, in addition to measuring the parent compound, also measured other metabolites, which may be pharmacologically active. Javaid et al. (1980) reported that the "neuroleptic activity" in plasma measured by radioreceptor assay was seven to eight times greater than the plasma FPZ levels measured by the GLC method. Harris et al. (1982) compared the plasma levels measured by radioreceptor assay and HPLC (high performance liquid chromatography) and found that the
58 receptor assay was able to detect FPZ in most samples, but H P L C failed to do so. Both reports emphasized the importance of active metabolites in plasma level studies. If the plasma levels are high, as in the patients receiving oral FPZ-H, the G L C method can be useful clinically (Dysken et al. 1981), since the total neuroleptic activity is proportional to the concentration of the parent c o m p o u n d (Javaid et al. 1980; and Harris et al. 1982). However, in patients receiving F P Z - D injections, the plasma level is extremely low and the G L C method, which only measures the parent compound, is not sufficiently sensitive. Therefore, the radioreceptor or radioimmunoassay method would be more suitable for the measurement of FPZ in this situation.
Acknowledgements. This work was done at the Illinois State Psychiatric Institute, Chicago IL and presented in part at the Third World Congress of Biological Psychiatry, Stockholm, Sweden, June 28-July 3, 1981 References
Curry SH, Whelpton R, deSchepper P J, Vrancks S, Schiff AA (1978) Plasma fluphenazine concentrations after injection of long-acting esters. Lancet i:121%1218 Curry SH, Whelpton R, deSchepper PJ, Vrancks S, Schiff AA (1979) Kinetics of fluphenazine after fluphenazine dihydrochloride, enanthate and decanoate administration to man. Br J Clin Pharmacol 7:325-331 Dekirmenjian H, Javaid JI, Duslak B, Davis JM (1978) Determination of antipsychotic drugs by gas-liquid chromatography with
a nitrogen detector using a simple acetylation technique. J Chromatogr 160: 291-296 Dysken MW, Javaid JI, Chang SS, Schaffer C, Shahid A, Davis JM (1981) Fluphenazine pharmacokinetics and therapeutic response. Psychopharmacology 73: 205-210 Franklin M, Wiles DH, Harvey DJ (1978) Sensitive gas chromatographic determination in human plasma. Clin Chem 24:41-44 Harris PQ, Friedman MJ, Cohen BM, Cooper TB (1982) Fluphenazine blood levels and clinical response. Biol Psychiatry 17:1123-1130 Javaid JI, Pandey GN, Duslak B, Hu H-Y, Davis JM (1980) Measurement of neuroleptic concentration by GLC and radioreceptor assay. Commun Psychopharmacol 4: 467-475 Javaid JI, Dekirmenjian H, Liskevych U, Lin R-L, David JM (1981) Fluphenazine determination in human plasma by a sensitive gas chromatographic method using nitrogen detector. J Chromatogr Sci 19:439-443 Nasrallah HA, Rivera-Calimlin L, Rogol AD et al. (1978) Fluphenazine decanoate: Plasma concentrations and clinical response, Psychopharmacol Bull 14:4~47 Tune LE, Creese I, Coyle JT, Pearlson G, Synder SH (1980) Low neuroleptic serum levels in patients receiving fluphenazine decanoate. Am J Psychiatry 137:80-82 Whelpton R, Curry SH (/976) Methods for study of fluphenazine kinetics in man. J Pharm Pharmacol 28:869-873 Wiles DH, Franklin M (1978) Radioimmunoassay for fluphenazine in human plasma. Br J Clin Pharmacol 5:265-268 Wiles DH, Gelder MG (1979) Plasma fluphenazine levels by radioimmunoassay in schizophrenic patients with depot injection of fluphenazine decanoate. Br J Clin Pharmacol 6:565-570 Received March 21, 1984; Final version March 12, 1985
