The WHO analgesic ladder has been extensively field tested (WHO, 1986; Walker et al., 1988; Schug et al., 1992) and is now considered to be a viable model for the pharmacological treatment of cancer pain. The vast majority of patients eventually reach step 3 of the analgesic ladder which involves the prescription of potent μ receptor opioid agonists (morphine is the drug of choice at this step) and a range of adjuvant medications. While various formulations of morphine are available, modified release formulations are being prescribed with increasing frequency because of convenience, continuous analgesia and possibly a lower incidence and severity of morphine related side-effects.
Recently, a new sustained release morphine formulation, called either Kapanol™ (Glaxo Wellcome group of companies) or KadianTM (F.H. Faulding and Co. Ltd), hereafter referred to as Kapanol, has been developed. Kapanol consists of small polymer coated pellets contained in a capsule. Compared to MS Contin, Kapanol exhibited significantly less fluctuation in plasma morphine concentrations throughout a dosing interval, lower maximum plasma morphine concentration (Cmax), longer Tmax (time that Cmax occurs) and time that the plasma morphine concentration ≥75% of the Cmax value (Time≥0.75 Cmax) following 12-h administration of both modified release formulations (Gourlay et al., 1994). The Time≥0.75 Cmax parameter is an index of the control the formulation exerts over the morphine release rate (from either individual pellets or tablets) and hence represents a measure of the effectiveness of the respective technologies used to modify the release characteristics of morphine in the two formulations.
Three factors, namely, the nature of the core, nature of the polymer coat, and the thickness of the coat collectively control the rate of release of morphine from each pellet. Each Kapanol capsule contains multiple pellets (for example, a 100-mg Kapanol capsule contains, on average 300 pellets), which has the effect of providing more controlled and uniform release characteristics. The control exerted over morphine release from the pellets was so precise that the possibility of one Kapanol dose per 24 h effectively controlling cancer pain was considered (Gourlay et al., 1994). This communication presents the pharmacokinetic and pharmacodynamic results of a randomised, double-blind, cross-over comparison of one Kapanol dose per 24 h compared to 12-h MS Contin in the treatment of moderate to severe cancer pain.
Twenty-four patients (15 male and nine female) with moderate to severe cancer pain completed a randomised, double-blind, double-dummy, cross-over study of once a day Kapanol compared to 12-h MS Contin. The two treatments were:
2.1. Ms Contin phase
Opaque capsules containing MS Contin were taken at 10:00 h and 22:00 h, and placebo capsules to match Kapanol capsules also at 10:00 h and 22:00 h.
2.2. Kapanol phase
Kapanol capsules were taken once a day at 10:00 h and placebo capsules to match Kapanol were taken at 22:00 h, and placebo opaque capsules matching those in which MS Contin was administered were taken at 10:00 h and 22:00 h. The active Kapanol dose was administered at 10:00 h to allow the maximum time for patient observation during day-light hours.
It was determined that 24 patients would be sufficient to provide 90% power, at a type 1 error rate of 0.05, to detect a 1.5-h difference in the time for which blood morphine concentrations were ≥75% of Cmax. The mean±SD (range) age and weight for the 24 evaluable patients were 65±13.2 (41–89) years and 69.2±14.5 (37–101) kg, respectively. Thirty-seven patients were enrolled and 29 patients were randomised into the study; four of the randomised patients were withdrawn during treatment due to disease progression (2 patients) and protocol violation (2 patients) while one patient who completed the study was withdrawn due to a protocol violation. The most common types of cancer in this study were breast (6 patients), prostate (6 patients), colon (5 patients), lung (4 patients), lymphatic (3 patients) and gastric and liver (2 patients each).
The inclusion criteria consisted of any adult patient with documented (pathological or radiological) evidence of cancer requiring the prescription of oral morphine at a dose of at least 40 mg per 24 h to control moderate to severe pain. Exclusion criteria were significant abnormalities in hepatic, renal, haematological, or pulmonary function and any gastro-intestinal pathology or surgery which could influence the absorption of morphine from either modified release formulation. Patients with a history of drug seeking behaviour or females who were either pregnant, lactating or using inadequate contraception were also excluded because of the investigational nature of Kapanol. Other exclusion criteria included patients undergoing chemotherapy and/or radiotherapy (either during periods 1 or 2, or 1 week before entry into the study), patients who were unable to complete daily diaries or comply with the protocol, an ECOG performance status >3 (0=fully active, 1=restricted in physically strenuous activity but ambulant and able to carry out light work, 2=ambulant and capable of self care but unable to carry out light work, 3=capable of only limited self care and confined to bed or chair >50% of waking hours, 4=completely disabled), known hypersensitivity to morphine, intractable vomiting or patients who could not swallow capsules whole.
The morphine dose was optimised (using an immediate release solution formulation) to provide the most favourable balance between analgesia and side-effects for each patient during the lead-in period. Patients recorded, in their daily diary, the administration time of all medication taken together with assessments of the efficacy (pain intensity, pain relief and quality of sleep) and side-effects (nausea/vomiting, constipation, sedation, confusion and loss of appetite) of the 4-h morphine regimen. When their 24-h morphine dose of solution was constant for at least two consecutive days (with no more than two rescue doses of morphine solution per 24 h), the patient was then randomised to either the Kapanol or MS Contin formulations in period 1 (which lasted for 7±1 days). The 24-h morphine dose used during periods 1 and 2 was the closest dose that could be administered using the various dose strengths of the two modified release formulations available (Kapanol; 20 mg, 50 mg and 100 mg: MS Contin; 10 mg, 30 mg, 60 mg and 100 mg) thereby preserving the double-blind, double-dummy nature of the study. Half of the patients had the Kapanol phase first and half had the MS Contin phase first (period 1). Patients changed to the alternative treatment in period 2. The MS Contin tablets were placed inside opaque capsules as placebo tablets to match MS Contin dose strengths were unavailable. Prestudy testing (11 healthy volunteers completed the bioequivalency study) established that encapsulated MS Contin tablets (0–36-h area under curve following a 60 mg morphine dose was 135±25.1 ng.h/ml) were bioequivalent to unencapsulated MS Contin tablets (equivalent area under curve was 143.6±36.1 ng.h/ml, unpublished data, report HL14700, F.H. Faulding and Co. Limited, Adelaide, Australia). In addition, the Cmax (mean values of 15.3 ng/ml and 15.5 ng/ml for MS Contin and encapsulated MS Contin respectively) and Tmax values (mean values of 2.3 h and 2.6 h for MS Contin and encapsulated MS Contin respectively) for the encapsulated and unencapsulated tablets were not significantly different from each other. Patients recorded the extent of analgesia and morphine related side-effects in their diary as previously described.
On day 7, patients were admitted to the Pain Management Unit and frequent blood samples (hourly for the first 12 h and 2 hourly thereafter) were collected for a total of 24 h following the 10:00 h dose via an indwelling centrally directed intravenous catheter usually in the antecubital fossa. Trough (10:00 h) blood samples were collected for two days prior to day 7 and assayed for morphine concentration to ensure patients were at steady state. Meal times for lunch and dinner on day 7 were controlled and the composition of the meals was identical in periods 1 and 2 for individual patients. Visual analogue pain scores (VAS) and the extent of pain relief were recorded at the time of blood sample collection, and a global assessment of efficacy by both patient (response assessed as either very good, good, fair or poor) and investigator (response assessed as either marked, moderate, minimal or no efficacy) was made at the completion of this phase. Period 2 immediately followed for a further seven days (first dose of the new medication at 10:00 h on day 8) and this period was identical to period 1 except the patients took the same 24-h morphine dose but via the alternative formulation. Patient preference (assessed as either treatment 1, treatment 2, both (equally) or neither treatment) with respect to formulation was obtained immediately after the completion of period 2. Rescue medication throughout periods 1 and 2 consisted of oral dextromoramide (5 mg tablets) as this opioid is frequently used for breakthrough pain and did not interfere with the morphine or morphine metabolite assay (Gourlay et al., 1987).
The primary pharmacodynamic efficacy variables were the percentage of patients taking rescue analgesia, the time to the first rescue dose on day 7 (during periods 1 and 2) and the amount of rescue analgesia taken on day 7. Secondary efficacy variables were based on VAS (10 cm, 0=no pain to 100=worst pain imaginable) and 4-point verbal rating scales (VRS) for pain intensity and pain control. These assessments were undertaken immediately before the first dose of study medication (i.e., day 1 of each period) and also after the first dose of rescue medication on day 7 of periods 1 and 2. In addition, these assessments were performed prior to every blood sample collected in the Pain Management Unit on day 7 (periods 1 and 2), every morning (at home) during periods 1 and 2, and at 2, 4, 8 and 12 h post-dose on day 1 of periods 1 and 2 (at home). A VRS relating to quality of sleep (0=very good sleep to 4=no sleep) was recorded every morning of periods 1 and 2 (at home), and graded assessments (usually: none, mild, moderate, severe or extreme (in some cases)) of the previously mentioned morphine related side-effects were recorded every evening during periods 1 and 2 (at home). The patient treatment code was not broken until the completion of the 24th evaluable patient.
Morphine in plasma samples was quantitated using the sensitive and specific method of high performance liquid chromatography (HPLC) with electro-chemical detection with a lower level of detection of 1 ng/ml (Todd et al., 1982). The standard curve was linear from 1 ng/ml to 80 ng/ml and quality control samples were included at three concentrations in every assay run. Plasma samples from the first eight evaluable patients were further analysed to quantitate the concentrations of the two glucuronide metabolites (i.e., morphine-3-glucuronide (M3G) and morphine-6-glucuronide (M6G)) by HPLC with u.v. detection (Milne et al., 1991). The lower level of quantitation and linear range for M3G and M6G were 40 ng/ml, 40–2000 ng/ml and 20 ng/ml, 20–500 ng /ml respectively. The quantitation of morphine in plasma samples was undertaken by Harris Laboratories (Lincoln, Nebraska, USA) while morphine metabolite quantitation was performed by the Analytical Laboratory, Faulding Drug Studies Unit (Adelaide, Australia). The following pharmacokinetic parameters were calculated from plasma morphine (or metabolite) concentration-time data: Cmax (ng/ml), represents the maximum morphine concentration in the dosing interval (dose normalised to 100 mg); Cmin (ng/ml), represents the minimum morphine concentration in the dosing interval (dose normalised to 100 mg); Tmax (h), represents the time associated with Cmax in the dosing interval; area under the plasma morphine concentration-time curve (AUC, ng.h/ml), represents the area under the plasma morphine concentration-time curve and is calculated using the trapezoidal rule (dose normalised to 100 mg); fluctuation in plasma morphine concentration throughout the 24-h dosing interval ((Cmax -Cmin)/Css where Css is the average concentration calculated as AUC/24 (dosing interval)); and the time the plasma morphine concentration was ≥75% Cmax (an index of the control the formulation exerts over the morphine release rate; the greater the value, the greater the level of control, h). Statistical analysis involved ANOVA, chi-squared tests and McNamar's test where appropriate, and a level of significance of at least P<0.05 was required to reject the null hypothesis.
3.1. Pharmacokinetic aspects
The patients were considered to be at steady state with respect to plasma morphine concentration because the slopes of the curves of plasma morphine concentrations obtained from the trough blood samples (10:00 h, 24:00 h samples on day 7 and the 10:00 h samples on days 5 and 6 during periods 1 and 2) as a function of time were not significantly different from zero. The mean±SD (range) morphine dose was 199±275 (40–1200) mg per 24 h for the 24 evaluable patients.
Fig. 1 shows the dose normalised (to 100 mg) plasma morphine concentration (ng/ml) as a function of time (hours) after the 10:00 h dose at steady state (i.e., day 7) for both Kapanol and MS Contin for three individual patients. Panel A is from a patient where the fluctuation of plasma morphine concentration is lower for Kapanol than for MS Contin (Kapanol=0.7; MS Contin=1.9), panel B is from a patient with similar fluctuation for both formulations (Kapanol=1.6; MS Contin=1.8) while panel C is from a patient with higher fluctuation for Kapanol than for MS Contin (Kapanol=2.0; MS Contin=1.3). Fig. 1 also shows both the plasma M3G and M6G concentrations (ng/ml) as a function of time in these patients for the Kapanol phase.
Fig. 2 shows the mean±SD dose normalised (to 100 mg) plasma morphine concentrations (log scale) as a function of time (hours) for both Kapanol and MS Contin (same symbols as in Fig. 1) in the 24 evaluable patients as well as the mean visual analogue pain scores for the Kapanol and MS Contin phases of the study. The variability in the dose normalised plasma morphine concentrations is relatively small with a coefficient of variation of approximately 50% at the various sampling times. The curves from the individual patients (Fig. 1) are similar to the mean curve (Fig. 2) except for minor differences in Tmax for both formulations. However, the individual curves (Fig. 1) shows patients exhibiting a range of fluctuations in plasma morphine concentration for the two modified release formulations.
A number of pharmacokinetic parameters were derived from the plasma morphine concentration-time curves from individual patients and the mean±SD values for these parameters are given in Table 1. While the Cmax values during the sampling interval were not significantly different between the two formulations, the Tmax values were significantly longer for Kapanol. Further, the Cmin values were significantly higher for Kapanol compared to MS Contin which resulted in a lower extent of fluctuation in plasma morphine concentration for Kapanol throughout the dosing interval. The time the plasma morphine concentration equalled or exceeded 75% of Cmax was significantly longer for Kapanol compared to MS Contin which reflects the greater control the Kapanol formulation exerts over the morphine release rate compared to that shown by MS Contin. The AUC values were not significantly different between the two formulations indicating that the extent of absorption of morphine from Kapanol administered 24 hourly was similar to that seen following 12-h MS Contin for the same 24-h morphine dose.
A subset (n=8) of patient samples were further analysed to quantitate the amounts of morphine metabolites, M3G and M6G. Fig. 1 also shows the glucuronide concentrations in the same individual patients for the Kapanol phase (MS Contin glucuronide concentrations have been omitted for clarity). It is apparent that there is a congruence between the shapes of plasma morphine and metabolite concentration-time profiles in individual patients. A similar congruence was observed between morphine and metabolite concentrations in the MS Contin phase of the study. The M3G and M6G areas under curve (AUC) were similar for both formulations (mean values of 21 991 and 21 604 ng.h/ml for M3G and 3303 and 3284 ng.h/ml for M6G for Kapanol and MS Contin phases respectively) as were the molar ratios for M3G:morphine (34.5 and 35.0 for Kapanol and MS Contin respectively) and M6G:morphine (5.1 and 5.3 for Kapanol and MS Contin respectively).
3.2. Pharmacodynamic aspects
A number of pharmacodynamic parameters were measured throughout the blinded periods of the study (see Methods section). The primary pharmacodynamic efficacy variables were the percentage of patients taking rescue medication and the time to the first rescue dose, as shown in Table 2. Patients in both treatment groups took similar amounts of rescue medication which approximated to 10% of the 24-h morphine dose. Notwithstanding some difference in potency between morphine and dextromoramide (Gourlay et al., 1987), it is reasonable to attribute the observed pharmacodynamic effects essentially to morphine. Approximately half of the patients took rescue medication on day 7 during both blinded periods of the study (42% for Kapanol versus 54% for MS Contin, P>0.05). The difference in the percentage of patients taking rescue analgesia in the MS Contin phase relative to the Kapanol phase was 12% and the confidence interval for that difference was -9% to +34%. Of the 24 evaluable patients, 10 patients took no rescue analgesia in either phase, nine patients took rescue analgesia in both phases, while five patients took rescue analgesia in only one of the two phases (four patients took rescue in the MS Contin phase only and one patient in the Kapanol phase). The mean time to first rescue medication in those patients who took rescue medication was 7.5 h and 5.6 h for Kapanol and MS Contin respectively (P>0.05). There was no significant difference in the amount of rescue medication taken on day 7 between the two formulations (Table 2).
Fig. 2 also shows the mean VAS (mm) for the Kapanol and MS Contin phases superimposed on the mean plasma morphine concentration (ng/ml) as a function of time following the 10:00 h dose at steady state (day 7) for the 24 evaluable patients. It is evident that the pain scores were low throughout the dosing interval on day 7 and that generally there was an inverse relationship between plasma morphine concentration and VAS scores; that is, when plasma morphine concentrations increased, VAS scores decreased and vice versa. There were no significant differences in VAS pain scores between the Kapanol and MS Contin phases at any of the time points shown in Fig. 2, nor in either peak or trough VAS pain scores for each formulation.
There were a number of other secondary pharmacodynamic variables measured during the blinded periods of the study as shown in Table 3. The various parameters in this table were usually assessed a number of times in a single day or over a number of days during each treatment period. The values given in the table are for those time points at which the magnitude of the difference between Kapanol and MS Contin phases was the greatest. There were no significant differences between the Kapanol and MS Contin treatments in any of the following parameters: verbal rating scale (VRS) for pain intensity at any of the sampling times on day 7, VRS for pain control at the same times indicated above, VAS pain scores, morning VRS pain intensity, morning VRS pain control, quality of sleep, incidence of nausea and vomiting, constipation, sedation and appetite assessed daily during periods 1 and 2. The only significant result was obtained on day 5 for the side-effect of confusion and this was considered to be a chance event as the assessments for this parameter on the other five days were not significantly different. The low numerical value for all of these pharmacodynamic parameters indicates that both formulations provided effective pain relief with a low level of side-effects.
Table 4 shows that there is no significant difference in patient global assessment and patient treatment preference between the two treatment phases. Table 5 shows there was no significant difference in investigator global assessment between the two treatment phases. The mean±SD pre- and post-treatment ECOG scores were 1.5±0.8 and 1.6±0.9 respectively. These changes were not significant nor were there any sequence effects.
A previous study found that 12-h Kapanol led to more constant plasma morphine concentrations than 12-h MS Contin (Gourlay et al., 1994). The shape of the plasma morphine concentration-time curve following Kapanol administration at 12-h intervals was very flat and showed only a small fluctuation in plasma morphine concentration (mean value of 0.6 for Kapanol versus 1.51 for MS Contin, P<0.05) and a prolonged time that the plasma morphine concentration ≥75% Cmax (mean value of 9.1 h for Kapanol compared to 3.1 h for MS Contin, P<0.05). Both the precise control over morphine release rate from Kapanol pellets and the large number of pellets in a dose contribute to the stable pharmacokinetic profile. The promising results of that study suggested the possibility that a single dose of Kapanol per 24 h could effectively treat moderate to severe cancer pain.
The results presented here were obtained using a study design incorporating the optimal and accepted elements of randomisation of patients to a double-blind, double-dummy, cross-over study where pharmacokinetic and pharmacodynamic parameters were determined at steady state (i.e., after seven days in periods 1 and 2). The morphine dose was optimised to provide the most favourable balance between pain relief and side-effects in the lead -in period. The morphine concentration data were dose normalised (to a nominal dose of 100 mg) prior to the calculation of pharmacokinetic parameters. The relatively small coefficient of variation of approximately 50% in the morphine concentrations at the various sampling times on day 7 (in periods 1 and 2) indicates that this mathematical manipulation is appropriate in this case (Fig. 2).
Kapanol, administered as a single dose per 24 h, has a less variable pharmacokinetic profile (Figs. 1 and 2, Table 1) compared to MS Contin, despite the fact that the dosing interval for Kapanol (24 h) was twice that of MS Contin (12 h). Classical pharmacokinetic dogma indicates that for a given total dose administered over a defined time period (e.g., say 100 mg per 24 h), the fluctuation in plasma concentrations is greater as the dosing interval increases (Hull, 1991). Thus, while the fluctuation in plasma morphine concentrations following once a day Kapanol is greater than 12-h Kapanol (Gourlay et al., 1994), it is nevertheless smaller than 12-h MS Contin (Table 1). The dose independent pharmacokinetic parameters of Tmax and time ≥75% Cmax (Table 1) are similar to those obtained following 12-h Kapanol and are significantly greater than those obtained from the MS Contin phase.
Fig. 1 shows that the M3G and M6G concentrations have similar time profiles to that of morphine concentrations. Data from individual patients show this type of relationship more clearly than do mean patient data. There are marked similarities in AUC for the metabolite concentration-time curves and the respective metabolite:morphine molar ratios between the two modified release formulations.
The primary pharmacodynamic variables selected before the study commenced were the percentage of patients requiring rescue analgesia and the time the first rescue dose was taken on day 7. Approximately half of the patients took no rescue analgesia on day 7, while the dextromoramide dose in the remainder of patients was approximately 10% of the morphine dose in the lead-in period. Thus, the pharmacodynamic effects observed, particularly on day 7, can reasonably be ascribed to the morphine concentrations from the two modified release formulations. The extent of pain relief observed following either Kapanol or MS Contin was very good as shown by the low VAS pain scores (Fig. 2) and the low amount of rescue analgesia taken on day 7. A previous study established a relationship between blood morphine concentration and the extent of pain relief (Gourlay et al., 1986). While there is a similar relationship in this study, the strength of the relationship is not strong because the morphine dose had been optimised in all patients.
The remarkable feature about the extensive number of pharmacodynamic measurements was that there were no significant differences between the Kapanol and MS Contin phases in any of these parameters (Table 3). These parameters covered the extent of pain and pain relief and morphine related side-effects throughout periods 1 and 2, but particularly on day 7 of each period. Similarly, there were no significant differences in patient or investigator global assessments performed at the end of each period (Tables 4 and 5). Although more patients indicated a preference for MS Contin than did for Kapanol, this difference was not statistically significant.
We conclude that Kapanol administered once a day has a superior pharmacokinetic profile to MS Contin administered twice a day. However, both modified release formulations provide an equivalent extent of pain relief and similar patterns of morphine related side-effects. There was no significant difference in patient global assessments or treatment preferences between the two formulations in this randomised, double-blind, cross-over study in cancer patients.
The authors gratefully acknowledge the significant contribution of our research nursing staff. Financial support to conduct this study was provided by F.H. Faulding and Co. Limited.
Gourlay, G.K., Cherry, D.A. and Cousins, M.J., A comparative study of the efficacy and pharmacokinetics of oral methadone and morphine in the treatment of severe pain in patients with cancer, Pain, 25 (1986) 297–312.
Gourlay, G.K., Cousins, M.J. and Cherry, D.A., Drug therapy. In: G.D. Burrows, D. Elton, and G.V. Stanley (Eds.), Handbook of Chronic Pain Management, Elsevier, Amsterdam, 1987, pp. 163–192.
Gourlay, G.K., Plummer, J.L., Cherry, D.A. and Onley, M.M., A comparison of Kapanol (a new sustained release morphine formulation), MST Continus, and morphine solution in cancer patients: pharmacokinetic aspects of morphine and morphine metabolites. In: G.F. Gebhart, D.L. Hammond and T.S. Jensen (Eds.), Progress in Pain Research and Management, Vol. 2, IASP Press, Seattle, 1994, pp. 631–643.
Hull, C.J., Pharmacokinetics for Anaesthesia, Butterworth Heinemann, Oxford, 1991, pp. 168–169.
Milne, R.W., Nation, R.L., Reynolds, G.D., Somogyi, A.A. and Van Crugten, J.T., High performance liquid chromatographic determination of morphine and its 3- and 6-glucuronide metabolites: improvements to the method and application to stability studies, J. Chromatogr., 565 (1991) 457–464.
Schug, S.A., Zech, D., Grond, S., Jung, H., Meuser, T. and Stobbe. B., A long term survey of morphine in cancer pain patients, J. Pain Symptom Manage., 7 (1992) 259–266.
Todd, R.D., Muldoon, S.M. and Watson, R.L., Determination of morphine in cerebrospinal fluid and plasma by high-performance liquid chromatography with electrochemical detection, J. Chromatog., 232 (1982) 101–110.
Walker, V.A., Hoskin, P.J., Hanks, G.W. and White, I.D., Evaluation of WHO analgesic guidelines for cancer pain in a hospital based palliative care unit, J. Pain Symptom Manage., 3 (1988) 145–149.
WHO Cancer Pain Relief, WHO, Geneva, 1986.