Validation Study of UPLC Method for Determination of Morphine, Ropivacaïne and Ziconotide in Combination for Intrathecal Analgesia

The management of cancer is not limited to the treatment of the disease but must also consider patient pain. In 15% to 20% of cancer patients, conventional analgesic therapy recommended by WHO fails to relieve pain [1]. In France, Intrathecal Analgesia (IA) is recommended by French Society of Anesthesia and Intensive Care Medicine (SFAR) in case of refractory chronic cancer pain, as well as in patients with treatment-related adverse effects impairing patient’s quality of life [2]. The administration of anesthetic drugs in the central nervous system can achieve similar or better analgesic results and can be easier to use for clinicians [3-5].

The administration of these analgesic drugs by intrathecal route is done by using a pump (an internal implantable or external intrathecal pump), connected with a tunneled catheter implanted in the intrathecal area [6,7]. The implantable pump has a reservoir which has to be refilled. External pumps have a bag containing the anesthetic substances that needs to be regularly replaced. This refill is prepared in a controlled atmosphere area under strict aseptic conditions in a vertical laminar flow cabinet by manipulators trained on these preparations to be used internally. Commercial pharmaceutical products are used for preparation.

Over the years, various molecules have been used in combination for IA. Morphine remains the reference in case of strong to intense pain. Ropivacaïne or bupivacaine are used to strengthen morphine effects by blocking nerve conduction in the sodium channels. Ziconotide is a small ω-conotoxine peptide isolated from a marine cone snail, Conus magus and acts as a selective blocker of voltage-sensitive calcium channels of N-type [8-11]. Due to its narrow therapeutic range and its potentially serious adverse effects (psychiatric and nervous system disorders), this drug is only indicated for the treatment of severe, chronic pain in adults who require intrathecal (IT) analgesia [12,13]. Finally, clonidine, sufentanil, ketamine can also be used, although these molecules are not present in the last International recommendations.

The Cancerology Center Paul Papin protocol includes an association of morphine, ropivacaïne and ziconotide in Intrathecal preparations (PIT). Concentrations of each analgesic agent vary from one preparation to another because they are adapted for each patient’s pain evaluated by the physician before the prescription [14]. Concentrations mean and range are listed in the Table 1.

Table 1: Mean concentrations of morphine, ropivacaïne and ziconotide in the PIT.

Dosage errors most likely explain a significant percentage of the side effects [15,16] caused by ziconotide. Ziconotide is a particularly powerful drug. Its therapeutic index is low and adverse effects may occur at low doses (1 μg/day) [8,14,17]. The accuracy of the injected dose is therefore vital and no other clinical study has checked concentrations of ziconotide really infused.

To secure the preparations, a systematic analytical control must be performed before release. Over time, several authors have studied the ziconotide dosage in combination with morphine and ropivacaïne by using methods of liquid chromatography combined with ultra-violet detection, although ziconotide concentrations were higher than those that we are actually using (25 μg/mL) [18-20].

The aim of this study was to develop and validate a new method of determination of ziconotide, morphine and ropivacaïne using ultra-performance liquid chromatography (UPLC) combined with an ultraviolet detector in order to control our preparation according to national and international recommendations [21-23].

Analytes and reagents

For development and validation of the method the commercial pharmaceutical products were used. Morphine (CDM Lavoisier, La Chaussée-Saint Victor, France) is available under a ready-to-use intravenous and intrathecal formulation of morphine sulfate 500 mg/10 mL including the excipients sodium chloride, hydrochloric acid and water for injection.

Ropivacaïne (Fresenius Kabi, Sèvres, France) is supplied in 20 mL bottles containing 200 mg and the excipients sodium chloride, hydrochloric acid, sodium hydroxide and water for injection. Ziconotide (Prialt^®, Eisai SAS, Paris La Défense, France) is provided in 100 μg/1 mL bottles, including the excipients methionine, sodium chloride, hydrochloric acid, sodium hydroxide and water for injection. The three drugs are mixed together, alone or with a saline solution of 0.9% as excipient (Fresenius Kabi, Sèvres, France), before refilling the intrathecal pump of 40 mL.

Acetonitrile and trifluoroacetic acid were supplied by VWR (Fontenay aux Roses, France) and ultrapure water was obtained from Elga Purelab DV25 system. Solvents are HPLC-MS/MS grade to avoid impurities like metal ions in the mobile phase.

Chromatographic and UV detection conditions

Analyses were conducted using an UPLC Acquity^® H-Class^® interfaced to a photodiode array detector (PDA) (Waters, Guyancourt, France). The column used was an Acquity UPLC^® BEH C18 1.7 μm 2.1 × 50 mm heated at 50°C and the mobile phase was a gradient of ultrapure water with 0.1% trifluoroacetic acid and acetonitrile with 0.1% trifluoroacetic acid. Two liquid chromatography (LC) methods were developed due to concentration range difference between analytes. Respective gradients are presented in Tables 2 and 3. For morphine and ropivacaïne analysis, 0.3 μL were injected into the column whereas 10 μL were injected for ziconotide. Detection was monitored at 285 nm for morphine, at 230 nm for ropivacaïne and at 200 nm for ziconotide. In these conditions, the retention times were of 0.62 minutes, 1.99 minutes and 1.62 minutes, respectively.

Table 2: Gradient used for morphine and ropivacaïne LC analysis.

Table 3: Gradient used for ziconotide LC analysis.

Standards and quality controls (QC) preparations

Standards of morphine (0.1 to 32 mg/mL) and ropivacaïne (0.312 to 10 mg/mL) were prepared by serial dilution of the pharmaceutical product with Nacl 0.9% in glass vials. Ziconotide standards (0.1 to 4 μg/ mL) were individually prepared by diluting the commercial solution in polypropylene vials to avoid wrong results when using serial dilution.

Quality controls (QC) containing a mean concentration of the three active substances (Morphine 4 mg/mL, Ropivacaïne 8 mg/mL and ziconotide 1 μg/mL) in NaCl 0.9% were prepared and analyzed during each validation run.

Validation procedure

The validation procedure was performed according to the US Food and Drug Administration (FDA) and Groupe d’Evaluation et de Recherche sur la Protection en Atmosphère Contrôlée (GERPAC) recommendations [22,23].

Selectivity: Selectivity with regard to excipients or other constituents of the mixture has been checked by analyzing individually three samples containing only the excipient (NaCl 0.9%) and the two other active substances in a mean concentration.

Calibration curve: The response of the instrument with regard to the concentration was studied. Concentrations were chosen on the basis of the expected range. For each analyte, six calibration standards and one zero standard have been analyzed during 3 runs (N=3), performed on three days by different manipulators.

Calibration curves were using a linear regression of the calibration points excluding the zero standards. Weighting and adequacy of the regression model was determined by statistical analysis (Levene’s test and lack-of fit test) according to the FDA recommendations [22]. The simplest model was used and between-run accuracy and precision of the standards were calculated.

QC within-run and between-run accuracy and precision: The accuracy and precision of the methods were evaluated by analyzing six individually prepared QC during each run. Within-run accuracy and precision were determined without reagent or manipulator variability in a single run (n=6) whereas between-run precision and accuracy were determined on three runs (total: n=18) performed in different days and involving several manipulators.

Accuracy was expressed as relative error (RE %) compared with nominal values and precision (repeatability) was determined by calculating the relative standard deviation (RSD %) of the measured QC concentration. Limits of acceptable were fixed at ±15% deviation from the expected for RSD and RE.

Carryover: Blanks containing 0.9% sodium chloride without analyte were injected after three high concentration samples (QC samples) in order to evaluate carry over for each analyte. This injection sequence was repeated 3 times.

Forced degradation study: As far as it was possible, a degradation of around 20% of the active substance was carried out by treating the analyte diluted in NaCl 0.9% with 0.5% hydrochloric acid followed by neutralization with 0.5% of sodium hydroxide. Each degradation product was identified by its retention time relative to the main analyte and mass balance was determined.

Selectivity

No interfering peaks were detected at the specific retention times of the three analytes of interest in the tested samples. The absence of interference by 0.9% NaCl and the two other active substances of the mixture have been validated for the three analytes. Thus, the matrix effect was not studied.

Response function analysis

Morphine and ziconotide, equality of variances was rejected and calibration curves were linear over the studied concentration ranges with 1/x regression. For ropivacaïne, homogeneity of variances was accepted according to the Levene’s Test (calculated F=2.932<theoretical F=3.106, IC 95%), so the linear regression was carried out without weighting. Calibrators calculated concentrations and between-run accuracy and precision are presented in Table 4. Acceptance criteria were met with absolute RE and RSD <15% for the three analytes. The regression coefficient (R²) for each calibration curve was >0.996.

Table 4: Calibrators between-run accuracy and precision.

QC within-run and between run accuracy and precision

Within-run and between-run accuracy and precision were determined with QC samples as described in the validation procedure section. Within-run precision ranged between -7.0% and 2.4% for ziconotide, -0.7% and -1.6% for ropivacaïne, -0.2% and -0.1% for morphine. Within-run and between accuracy and precision met the acceptance criteria with RSD <15% and inaccuracy ranged from -15% to +15% for the three studied molecules (Table 5).

Table 5: Within-run and between-run accuracy and precision obtained for QC samples.

Carryover

No peak was detected in the blank series, meaning that no intersample contamination occurred for ziconotide, morphine and ropivacaïne.

Degradation study

Ziconotide (1.3 μg/mL) was degraded by adding 0.5% HCl to the final concentration. The vial was opened during 3 hours in order to mimic a prolonged exposition to oxygen and allow oxidation reactions representing 20.54% of the initial compound was observed at a relative retention time of 0.95 minutes (Figure 1). The mass balance obtained was 100.1%, which proves that no other degradation product was found (Table 6). This degradation product has a specific retention time of 1.555 minutes and will be therefore visible in samples containing morphine and ropivacaïne.

Table 6: Degradation products of ziconotide.

Figure 1: Chromatogram of the degradation products for Ziconotide.

No degradation product was observed for morphine or ropivacaïne despite treatments with hydrochloric acid, hydrogen peroxide, sodium hydroxide or UV.

Application

After validation, the method has been routinely used for PIT preparation analysis. During the first month 248 PIT mixtures (including 224 with ziconotide) have been prepared and analyzed before release. Samples with concentrations out of range were diluted. The acceptance criteria were ± 10% for morphine and ropivacaïne and ±15% for ziconotide. The absolute RE means were between 2.1% and 3.2% for morphine, between 0.9% and 3.1% for ropivacaïne and between 5.6% and 9.3% for ziconotide (Table 7). One preparation has been rejected (absolute RE -22.80%).

Table 7: Absolute RE mean obtained for routinely analyzed PIT preparations.

Intrathecal Analgesia is in progression and our activity is increasing with a progression of 54% between 2012 and 2013. A method based on UPLC-UV to quantify admixture with morphine, ropivacaïne and ziconotide was developed and fully validated according to FDA recommendations. The calibration range was from 0.1 to 4 μg/mL for ziconotide, 0.316 to 10 mg/mL for ropivacaïne and 0.1 to 32 mg/mL for morphine. This method offers an interesting compromise in terms of analysis time and detection level compared with previously published methods [18,20]. This quantitative analysis limits dosage errors and overdosing, particularly for ziconotide. The method for ziconotide, morphine and ropivacaïne determination that we developed was validated, used in routine and can be consider as a stability indicating method. In order to determine the expiration date, a stability study will be carried out. Results may allow us to develop subcontracting activity which would consist in preparations shipment in other centers.