Method for detecting content of lamidetan hemisuccinate dihydrate in lamidetan tablet

Abstract

The invention belongs to the field of medicine crystal form quality control, and particularly relates to a quantitative method for a lamidipine hemisuccinate dihydrate crystal form D. The determination method has high specificity and accuracy for determination of the content of the lamidipine hemisuccinate dihydrate crystal form D in the lamidipine tablet. The content of the low-concentration crystal form D can be accurately quantified, and a research thought and a feasible method are provided for detecting the low-content crystal form in a medicinal preparation.

Description

Technical Field

[0001] The present invention relates to a method for quantitatively detecting lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets, and specifically relates to a method for rapidly and accurately quantitatively detecting the content of lamiditan hemisuccinate dihydrate crystal form D in a pharmaceutical preparation by using near-infrared spectroscopy technology, and belongs to the field of drug quality control. Background Art

[0002] Lamiditan tablets were developed by Eli Lilly and Company and were first approved for marketing by the U.S. FDA on October 11, 2019. The trade name is Reyvow. It is the first 5-HT1F receptor agonist approved for marketing. The drug is mainly used to treat acute migraine with or without aura in adults. The stable component of the preparation is lamiditan hemisuccinate. According to literature reports, there are 7 crystalline forms of lamiditan hemisuccinate, namely, crystal form A (anhydrate), B, C, D (dihydrate), E (dehydrate hydrate of crystal form D), F (trihydrate) and amorphous.

[0003] The thermodynamic stability of the anhydrous crystal form A of lamiditan hemisuccinate is close to that of the dihydrate crystal form D of lamiditan hemisuccinate. Crystal form D is not completely stable under low relative humidity conditions and may undergo crystal transformation. In order to ensure that the crystal form of the API is stable and reliable during the preparation process and stability study, the crystal form of the API in lamiditan tablets was studied in depth from the aspects of formulation design, process route determination, process parameter selection, material control, etc., according to the concept of quality by design (QbD), and multiple quantitative detection methods of crystal forms were developed. After comparison, near-infrared spectroscopy has the advantages of high sensitivity, rapidity and accuracy, and is very suitable for quality release detection and stability inspection of the dihydrate crystal form D of lamiditan hemisuccinate in this product. Near infrared spectroscopy is a spectral analysis technology that records the double frequency and combined frequency absorption of CH, NH, CO and OH bonds, and then analyzes their spectra through chemometric methods to perform qualitative and quantitative analysis of substances. It has the characteristics of rapid, accurate and non-destructive detection of samples. It can not only perform offline analysis, but also directly perform online process control; it can not only directly determine the active ingredients in raw materials and preparations, but also analyze certain physical and chemical properties of drugs such as water content, hydroxyl value, iodine value and acid value of fatty compounds. At present, no literature reports on the quantitative detection of lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets based on near infrared spectroscopy or other technologies have been found. Summary of the invention

[0004] The purpose of the present invention is to solve the problems of currently commonly used crystal form quantitative detection methods (such as powder X-ray diffraction method), such as poor differentiation of different crystal forms, no available characteristic peaks, long detection time, inability to monitor the production process in real time, complex finished product detection operation, poor sensitivity, etc. By establishing a linear regression between characteristic band spectral data and different crystal form contents, the lamiditan hemisuccinate dihydrate crystal form D (also referred to as crystal form D in the present invention) in lamiditan tablets is monitored.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets, the technical solution comprising the following steps:

[0007] (1) Preparation of reference substances

[0008] Several lamiditan preparation reference substances containing different concentrations of lamiditan hemisuccinate dihydrate crystalline form D were prepared as calibration set samples, and several lamiditan preparation reference substances containing different concentrations of lamiditan hemisuccinate dihydrate crystalline form D were prepared as validation set samples.

[0009] The lamiditan preparation reference substance also contains lamiditan hemisuccinate crystal form A (also referred to as crystal form A in the present invention); the content of crystal form D in the calibration set samples ranges from 0.00% to 4.00% (w / w), with a total of 18 reference substances; the content of crystal form D in the verification set samples ranges from 0.30% to 4.01% (w / w), with a total of 6 samples.

[0010] The lamiditan hemisuccinate crystal form A is lamiditan hemisuccinate anhydrate, and the crystal form D is lamiditan hemisuccinate dihydrate.

[0011] The calibration set samples and the validation set samples are obtained by mixing the API and the blank excipients. Before preparation, the API and the blank excipients must pass through a 100-mesh sieve to maintain consistent particle size, and must be mixed evenly before measurement. The API is selected from different crystal forms of lamiditan hemisuccinate, specifically lamiditan hemisuccinate crystal form A and / or lamiditan hemisuccinate dihydrate crystal form D.

[0012] The concentration range of the calibration set samples should not be too large or too small. An appropriate concentration range should be selected. The number of samples should be greater than or equal to 10. The more samples there are, the higher the accuracy. The concentration range of the validation set samples should be consistent with the concentration range of the calibration set samples.

[0013] (2) Near-infrared spectrum acquisition

[0014] The spectral information of the calibration set samples and the validation set samples was collected using a near-infrared spectrometer.

[0015] The specific parameters for collecting reference substances by the near-infrared spectrometer are as follows: the scanning range is 4000-12000 cm -1 , the resolution is 8 cm -1 , the number of scans is 64 times, repeated scanning is 3 times, and the average spectrum is taken; the measurement is carried out by the method of rotating integrating sphere, and each reference substance is about 200 mg.

[0016] (3) Pretreatment of the original spectrum

[0017] The obtained near-infrared spectra of the calibration set samples are pretreated, and the characteristic band range is selected according to the spectral differences between different crystal forms of the API and the blank excipients, and the spectral data of the calibration set samples at the characteristic wavenumbers are obtained.

[0018] The pretreatment method is a combination of standard normal variate transformation (SNV), first derivative and SG smoothing methods; the characteristic band range is 4899-6279 cm -1 .

[0019] (4) Model establishment

[0020] A quantitative model of spectral data and the content of different crystal forms is established.

[0021] The quantitative model is established by partial least squares regression (PLS) for the spectral data at the characteristic wavenumbers and the content of different crystal forms of ramelteon sesquisuccinate dihydrate crystal form D.

[0022] (5) Model evaluation

[0023] The correlation coefficient (R 2 ) of the calibration set, the root mean square error (RMSEC) and the root mean square error of cross-validation (RMSECV) are used to evaluate the performance of the model for the validation set samples, and the results of the methodology verification must meet the requirements.

[0024] The R 2 of the calibration set approaches 1, RMSEC < RMSECV, both are small and the numerical difference is not large; the methodology verification includes specificity, linearity, durability, precision (method repeatability and intermediate precision) and accuracy.

[0025] (6) Determination of the sample to be tested

[0026] The near-infrared spectrum of the ramelteon tablet sample with unknown content of ramelteon sesquisuccinate dihydrate crystal form D is measured at the characteristic wavenumbers, and the content of crystal form D of the sample to be tested is obtained according to the quantitative model established in step (4).

[0027] Compared with the prior art, the technical solution of the present invention has the following advantages:

[0028] (1) Compared with the powder X-ray method, the present invention has simple operation, fast detection speed and high sensitivity;

[0029] (2) Compared with the DSC method, the present invention can greatly eliminate the interference of free water and excipients on the detection of raw materials, and has a fast detection speed and high accuracy;

[0030] (3) The present invention provides a research idea and a feasible method for the identification and detection of low-content crystal forms in pharmaceutical preparations. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The original near-infrared spectrum and the pre-processed spectrum of the reference substance of lamiditan preparation;

[0032] Figure 2 It is the original near infrared spectra of lamiditan hemisuccinate dihydrate crystal form D, lamiditan hemisuccinate crystal form A and blank excipients;

[0033] Figure 3 is the calibration set linear regression model;

[0034] Figure 4 is the actual value-measured value curve of the validation set. DETAILED DESCRIPTION

[0035] The present invention is further described below by means of specific embodiments, but the protection scope of the present invention is not limited thereto.

[0036] Example 1: Method for determining the content of lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets

[0037] This embodiment relates to a quantitative method for the crystalline form of lamiditan hemisuccinate in lamiditan tablets, and the specific detection steps are as follows: 1. Preparation of calibration set and validation set reference substances

[0038] First, several lamiditan preparation reference substances (calibration set reference substances) containing different concentrations of lamiditan hemisuccinate dihydrate crystal form D need to be prepared for model establishment, and then several lamiditan preparation reference substances (validation set reference substances) containing different concentrations of lamiditan hemisuccinate dihydrate crystal form D need to be prepared to evaluate model performance. Before preparing the sample, the blank excipient, lamiditan hemisuccinate crystal form A and crystal form D need to be ground into powder and passed through a 100-mesh sieve to keep the particle size consistent as much as possible. The grinding in this operation process cannot cause the crystal form of the raw material to transform.

[0039] First, mother samples containing lamiditan hemisuccinate crystal form A and crystal form D were prepared according to the prescription ratio of lamiditan preparation, and then reference substances were prepared by stepwise dilution. The content (w / w) of crystal form D in the calibration set was 0.00% (4 samples), 0.30% (2 samples), 0.50% (2 samples), 0.75% (1 sample), 1.00% (2 samples), 1.50% (1 sample), 2.00% (2 samples), 2.50% (1 sample), 3.00% (1 sample) and 4.00% (2 samples), a total of 18 reference substances, and the content (w / w) of crystal form D in the verification set was 0.30%, 0.50%, 1.00%, 2.00%, 3.99% and 4.01%, a total of 6 samples.

[0040] 2. Near-infrared spectroscopy acquisition

[0041] Reference sample collection: Specific parameters are: scanning range is 4000~12000cm -1 , resolution 8cm -1 The scanning times were 64 times, and the measurement was carried out using a rotating integrating sphere. Each reference sample was scanned 3 times and the average spectrum was taken.

[0042] Each reference substance weighs about 200 mg and is placed in a special sample tray so that the sample is spread flat on the bottom of the sample tray and pressed into a flat surface. In order to reduce the influence of the homogeneity difference of the reference substance on the quantitative model, multiple samples can be taken for spectrum collection for each reference substance; in this embodiment, a total of 37 samples are taken from 18 calibration set reference substances for near-infrared spectrum collection, and at least one sample is taken for each calibration set reference substance.

[0043] 3. Raw spectrum processing

[0044] In the process of solid sample spectrum acquisition, the systematic error of near-infrared spectrum data acquisition is mainly affected by light scattering, sample particle size and effective path. Standard normal variable transformation can effectively eliminate the influence of scattering and particle size factors. The first-order derivative is suitable for eliminating baseline drift. Smoothing can eliminate noise and improve the spectral signal-to-noise ratio. Therefore, the spectrum is preprocessed accordingly before building the model to reduce the overlap between absorption peaks and make the effective information appear. Before building the model, the spectrum is preprocessed with a combination of first-order derivative + SG smoothing (17 points) + SNV, see Figure 1 According to the spectral differences between the blank excipient, lamiditan hemisuccinate crystal form A and crystal form D, the characteristic band range was selected to obtain the spectral data of the reference substances with different crystal content at the characteristic wavenumber. The characteristic band range was 4899-6279 cm -1 ,See Figure 2 .

[0045] 4. Build the model

[0046] Using TQ Analyst quantitative analysis software, first derivative + SG smoothing (17 points) + SNV preprocessing was performed on the calibration set samples in the spectral range of 4899 - 6279 cm -1 A quantitative model between the spectral data at characteristic wavenumbers and different contents of polymorph D was established by PLS.

[0047] 5. Model evaluation

[0048] The R of the established model 2 is 0.9968, RMSEC is 0.0889, RMSECV is 0.0970, R 2 is close to 1, RMSEC < RMSECV, and both are relatively small, indicating that the model has a good fit. See Figure 3 .

[0049] The established model was further subjected to methodological verification.

[0050] (1) Specificity

[0051] Based on near-infrared diffuse reflectance spectroscopy analysis technology and chemometric methods, a quantitative model was established. The determination coefficient R of this model 2 is 0.9968, RMSEC is 0.0889, RMSECV is 0.0970, R 2 is close to 1, RMSEC < RMSECV, and both are relatively small, indicating that the model has a good fit. It can be visually judged from the NIR actual value - measured value curve that the actual value and the measured value are basically the same, and the model can well identify the spectral characteristics of lamotrigine hemisuccinate dihydrate polymorph D.

[0052] (2) Linearity

[0053] The near-infrared spectral data obtained from the validation set samples were imported into the quantitative model, and the performance of the model was tested using the validation set samples. Its content range is consistent with the content range of the calibration set reference substances. The results show that the linear relationship between the actual value - measured value curve of the samples measured is good (R 2 = 0.9962). In the concentration range of 0.30% - 4.01%, the recovery rate is between 86.0% - 110.0%. Therefore, this PLS model is ideal. The results are shown in Table 1 below, as Figure 4 .

[0054] Table 1 Actual values, measured values and recovery rates of the samples measured

[0055] serial number Actual value (%) Measured value (%) Recovery rate (%) C-6 4.01 3.82 95.3 C-7 3.99 4.02 100.8 C-8 2.00 2.07 103.5 C-9 1.00 0.86 86.0 C-10 0.50 0.51 102.0 C-11 0.30 0.33 110.0

[0056] (3) Robustness

[0057] Three 3.00% concentration (w / w) Form D samples were prepared in parallel, and the experimental parameters were changed: vortex mixing time (10min, 13min, 16min) and scanning times (32 times, 64 times, 96 times), and near-infrared spectra were collected to calculate the relative standard deviation (RSD). The results showed that the RSD value was 0.93%, and the method robustness met the requirements. The results are shown in Table 2 below.

[0058] Table 2 Determination of the content of Form D in the durability test and the relative standard deviation (RSD)

[0059]

[0060] (4) Precision

[0061] Method reproducibility

[0062] Six 3.00% concentration crystal form D reference substances were prepared in parallel, near infrared spectra were collected, and relative standard deviation (RSD) was calculated. The results showed that the RSD value was 4.7%, which met the requirements. The results are shown in Table 3 below.

[0063] Table 3 Measured values ​​and relative standard deviation (RSD) of the content of Form D in the repeatability test of the method

[0064]

[0065] Intermediate precision

[0066] Different operators prepared 6 3.00% concentration crystal form D reference substances in parallel, collected near infrared spectra, and calculated the relative standard deviation (RSD). The results showed that the RSD value was 5.5%, which met the requirements. The results are shown in Table 4 below.

[0067] Table 4 Determination value and relative standard deviation (RSD) of the content of Form D in the intermediate precision test

[0068]

[0069] (5) Accuracy

[0070] Six 1.00% concentration of Form D reference substances were prepared in parallel, and near infrared spectra were collected. The recovery (%) and relative standard deviation (RSD) were calculated. The results showed that the recovery of the six parallel reference substances was between 88.0% and 111.0%, and the RSD was 9.0%, and the accuracy of the method met the requirements. The results are shown in Table 5 below.

[0071] Table 5 Measured value, recovery rate and relative standard deviation (RSD) of the content of Form D in the accuracy test

[0072]

[0073] The precision (RSD is 9.0%) and accuracy of the 1.00% concentration of Form D reference method both meet the limit of quantitation (LOQ) requirements, that is, the quantitation of this method is 1.00%; according to the relationship between the limit of quantitation and the limit of detection (LOD) (LOD = LOQ / 3), the detection limit of this quantitative model is 0.33%.

[0074] Conclusion: The results of the methodological validation for the content determination of lamiditan hemisuccinate dihydrate crystalline form D in lamiditan tablets showed that the specificity, linearity, durability, precision and accuracy of the quantitative model were good, and the present invention was suitable for the quantitative determination of the content of lamiditan hemisuccinate dihydrate crystalline form D in lamiditan tablets.

[0075] Example 2: Determination of the content of lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets

[0076] Determination method: Specific parameters are: scanning range is 4000~12000cm -1 , resolution 8cm -1 The scanning times were 64 times, and the measurement was carried out using a rotating integrating sphere. About 200 mg of each sample was taken, and the scanning was repeated 3 times to take the average spectrum.

[0077] Five batches of lamiditan tablet samples were taken and fully ground with an agate mortar, then passed through a 100-mesh sieve, and about 200 mg of the fine powder was taken for testing. The spectral data was imported into the model to calculate the content of lamiditan hemisuccinate dihydrate form D. The results are shown in Table 6.

[0078] Table 6 Results of the content of lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets

[0079] Sample No. Crystal form D content (%) Conclusion (LOD = 0.33%) Homemade 1 0.03 Not detected Homemade 2 0 Not detected Homemade 3 0 Not detected Homemade 4 0.12 Not detected Homemade 5 0.03 Not detected

[0080] Conclusion: The data showed that no lamiditan hemisuccinate dihydrate form D was detected in the five batches of lamiditan tablets; the quantitative determination method is suitable for the determination of lamiditan hemisuccinate dihydrate form D in lamiditan tablets.

[0081] The above embodiments are merely examples for clear explanation and are not limited to the protection scope of the present invention. All modifications or changes based on the basic concept of the present invention belong to the protection scope of the present invention.

Claims

1. A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets, characterized in that: The following steps are involved: (1) Preparation of reference substances: Several lamiditan preparation reference substances containing lamiditan hemisuccinate dihydrate crystalline form D at different concentrations were prepared as calibration set samples, and several lamiditan preparation reference substances containing lamiditan hemisuccinate dihydrate crystalline form D at different concentrations were prepared as validation set samples; (2) Near-infrared spectral acquisition: Fourier transform near-infrared spectrometer is used to collect spectral information of calibration set samples and validation set samples; (3) Processing the original spectrum: Preprocess the obtained near-infrared spectra of the calibration set samples, select the characteristic band range according to the spectral differences between different crystalline forms of the API and the blank excipients, and obtain the spectral data of the calibration set samples at the characteristic wavenumbers; (4) Model building A quantitative model for spectral data and different contents of Form D was established; (5) Determination of the sample to be tested The near infrared spectrum of a sample of lamiditan tablets with an unknown content of lamiditan hemisuccinate dihydrate crystalline form D is measured at a characteristic wavenumber, and the crystalline form D content of the sample to be tested is obtained according to the quantitative model established in step (4).

2. The quantitative detection method of lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets according to claim 1, characterized in that: In step (1), the lamiditan preparation reference substance also contains lamiditan hemisuccinate crystalline form A, and 10 or more calibration set samples are prepared, and the content of crystalline form D is in the range of 0 to 4% (w / w), including but not limited to 0.00%, 0.30%, 0.50%, 0.75%, 1.00%, 1.50%, 2.00%, 2.50%, 3.00% and 4.00% (w / w); 6 or more validation set samples are prepared, and the content of crystalline form D is in the range of 0 to 4% (w / w), including but not limited to 0.30%, 0.50%, 1.00%, 2.00%, 3.99% and 4.01% (w / w).

3. A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets according to any one of claims 1 to 2, characterized in that: In step (2), the specific parameters of the reference sample collected by the near infrared spectrometer include but are not limited to: the scanning range is 4000 to 12000 cm -1 , resolution 8cm -1 The scanning times were 64, the measurement was repeated 3 times, and the average spectrum was taken.

4. A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets according to any one of claims 1 to 3, characterized in that: In step (3), the preprocessing method is a combination of standard normal variable transformation, first-order derivative and SG smoothing method; the characteristic band range is 4899~6279cm -1 .

5. A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets according to any one of claims 1 to 4, characterized in that: In step (4), partial least squares regression (PLS) is used to establish a quantitative model of the spectral data at characteristic wavenumbers and different contents of form D.

6. A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets according to any one of claims 1 to 5, characterized in that: The method further comprises a model evaluation step after step (4), wherein the model evaluation uses the calibration set correlation coefficient (R 2 ), root mean square error (RMSEC), root mean square error of cross validation (RMSECV), and validation set samples to evaluate model performance.

7. A method for quantitatively detecting the crystalline form D of lamiditan hemisuccinate dihydrate in lamiditan tablets according to any one of claims 1 to 6, characterized in that: The lamiditan preparation reference substance used as the calibration set sample and the validation set sample in step (1) is obtained by mixing the raw material drug and the blank excipient. Before preparation, the raw material drug and the blank excipient are ground with an agate mortar and passed through a 100-mesh sieve. The raw material drug is selected from lamiditan hemisuccinate crystal form A and lamiditan hemisuccinate dihydrate crystal form D.

8. A method for quantitatively detecting the crystalline form D of lamiditan hemisuccinate dihydrate in lamiditan tablets according to any one of claims 1 to 7, characterized in that: The sample to be tested in step (5) needs to be ground in an agate mortar and passed through a 100-mesh sieve before testing.

9. A quantitative detection method for lamiditan hemisuccinate dihydrate crystal form D in lamiditan tablets according to any one of claims 1 to 8, characterized in that: In the steps (2) and (5), the near infrared spectra of the reference substance or the sample to be tested are collected by a rotating integrating sphere method.

Images