2024 Volume 47 Issue 5 Pages 878-885
The existence of substandard and falsified medicines threatens people’s health and causes economic losses as well as a loss of trust in medicines. As the distribution of pharmaceuticals becomes more globalized and the spread of substandard and falsified medicines continues worldwide, pharmaceutical security measures must be strengthened. To eradicate substandard and falsified medicines, our group is conducting fact-finding investigations of medicines distributed in lower middle-income countries (LMICs) and on the Internet. From the perspective of pharmaceutics, such as physical assessment of medicines, we are working to clarify the actual situation and develop methods to detect substandard and falsified medicines. We have collected substandard and falsified medicines distributed in LMICs and on the Internet and performed pharmacopoeial tests, mainly using HPLC, which is a basic analytic method. In addition to quality evaluation, we have evaluated the applicability of various analytic methods, including observation of pharmaceuticals using an electron microscope, Raman scattering analysis, near-IR spectroscopic analysis, chemical imaging, and X-ray computed tomography (CT) to detect substandard and falsified medicines, and we have clarified their limitations. We also developed a small-scale quality screening method using statistical techniques. We are engaged in the development of methods to monitor the distribution of illegal medicines and evolve research in forensic and policy science. These efforts will contribute to the eradication of substandard and falsified medicines. Herein, I describe our experience in the development of detection methods and elucidation of the pharmaceutical status of substandard and falsified medicines using novel technologies.
The problem of substandard and falsified medicines is an issue that must be solved on a global scale. WHO defines substandard medicines, also called “out of specification,” as authorized medical products that fail to meet their quality standards or specifications, or both. Falsified medical products are those with deliberately misrepresented or fraudulent identity, composition, or source.1) Substandard and falsified medicines are present in every region of the world. Their existence causes not only health damage but also economic damage, and the situation is becoming more serious with the globalization of supply chains.
The distribution of substandard and falsified medicines is increasing globally, especially in lower middle-income countries (LMICs), where the proportion of substandard and falsified medicines is as high as 10.5% of the pharmaceutical market.2) Pharmaceutical regulatory authorities, pharmaceutical companies, and researchers in LMICs are working to deter the distribution of substandard and falsified medicines,3) but the current situation regarding the distribution of pharmaceutical products requires further improvement. Japan also faces the threat of falsified medicines. Worryingly, substandard and falsified medicines that have already been identified only represent the tip of the iceberg.3)
To exploit the wishes of consumers and patients, the targets of falsification change daily, and falsified medicines have become more sophisticated. Additionally, the profits from falsified medicines are considered to be a source of funds for criminals, and there is concern that this will lead to further expansion of criminal activity in this area. As the distribution of pharmaceuticals becomes more globalized and the spread of substandard and falsified medicines continues worldwide, the development of methods to detect substandard and falsified medicines and deter their inappropriate distribution is needed to strengthen pharmaceutical security measures. Under these circumstances, I am working to strengthen pharmaceutical security measures with the aim of protecting patients from these risks, focusing on the development of detection technology (Fig. 1). To develop technology for the detection of substandard and falsified medicines, we analyzed pharmaceutical products that were actually distributed using a variety of methods, including observation that does not require equipment, non-destructive analysis that is possible even after analysis, and destructive analysis in which we mainly use HPLC. From the perspectives of evaluating the physical properties of pharmaceuticals, forensic science, and regulatory science, I aim to eliminate substandard and falsified medicines, the inappropriate distribution of medicines, and to meet the needs of this field. My goal is to realize the sustainability of medicines, deliver medicines with reliable quality, and ensure their proper use. This will ultimately lead to the proper use of medicines and greater health benefits from more effective and safer pharmaceutical treatment.
I am working to strengthen pharmaceutical security measures, focusing on the development of detection technology with the aim to deliver medicines with reliable quality, ensuring their proper use and realizing the sustainability of medicines.
My research on substandard and falsified medicines began with an investigation of the status of medicines available on the market in LMICs as well as proposals for improvement measures based on that investigation.4–7) To solve the problem of falsified medicines in Japan, our group also investigated the status of medicines distributed on the Internet.8,9) Our work clarified the fact that “health products” obtained via the Internet advertising a weight-loss effect, which claimed to only contain ingredients derived from natural products, actually contained the unapproved pharmaceutical ingredient sibutramine whose approval was withdrawn owing to serious side effects. Our findings called attention to the easy personal import of such health products.
Through the studies described above, it was possible to grasp the epidemiological situation regarding the problem of substandard and falsified medicines. However, the nature of low-quality and falsified medicines remained unclear. Moreover, techniques to detect substandard and falsified medicines, which is essential for deterring their distribution and preventing harm to health, had not yet been developed. This was the impetus for research on the development of methods to detect substandard and falsified medicines based on clarifying their pharmaceutical characteristics using modern technology. To accomplish this, it is essential to grasp the actual situation regarding medicines that are available on the market, which requires great effort. We continued to investigate the quality of medicines distributed in LMICs and personally imported medicines10–19) and to develop detection methods.20–30) In addition to demonstrating the applicability of spectroscopic analysis, some results obtained in these studies could be extremely valuable from the perspective of contributing to planning countermeasures that match the actual situation by analyzing currently available medicines that are substandard and falsified. The main research outcomes are described below.
To elucidate the pharmaceutical status of substandard medicines, we applied electron microscopy and X-ray CT in pharmaceutical analysis. We sought to visualize the formulation structure of pharmaceutical products with poor dissolution that are distributed in LMICs.7,13) We clarified that omeprazole capsules with severe dissolution failure lacked the enteric coating that should have been present in the granules inside the capsule.20) As a result of observing the intracapsular granules of enteric-coated omeprazole formulations with poor dissolution using an electron microscope, a smooth film was observed on the surface in the genuine innovator product (Fig. 2A). In contrast, there were no glossy surfaces and there were cracks found, among the intracapsular granules with poor dissolution that were distributed in an LMIC (Figs. 2B, C). Additionally, although there appeared to be a film coating, the granules were broken (Fig. 2D). In observing the granules inside the capsule of enteric-coated omeprazole using X-ray CT, we also found that the enteric coating, which can be seen in the genuine innovator product, was missing in some samples with poor dissolution distributed in LMICs. In the innovator product, the color of granules was uniform and a film coating was observed on the surface of the granule (Fig. 3A). In samples with poor dissolution distributed in LMICs, the color of granules within the capsules was not uniform. One capsule contained pale yellow granules with a film coating (Fig. 3B) and white granules without a film coating (Fig. 3C). In another capsule, we found white granule that appeared to have a film coating (Fig. 3D), white granules with no film coating and that were broken (Fig. 3E), and pale yellow granules without a film coating (Fig. 3F). Although differences in the components present on the surface of the granules inside the capsule of omeprazole tablets could be detected using Raman scattering analysis, X-ray CT clarified the cause of the difference in Raman spectra. Additionally, voids and electron-dense aggregates such as metal elements that were not observed in authentic candesartan tablets, could be seen in falsified tablets using X-ray CT.21) As a result, the existence of pharmaceutical defects that had been overlooked was revealed, making it possible to propose quality improvement measures based on scientific evidence. By accumulating this information and clarifying the relationship with falsified medicines, even when the authenticity of a product is unknown, this approach can be used to detect falsified medicines.
In the genuine innovator product (A), a smooth film was observed on the surface. In contrast, the three samples with poor dissolution that were distributed in a lower middle-income country (B–D) showed no gloss on the surface and it had cracks (B, C); although there appeared to be a film coating, the granules were broken (D).
A) Surface of granules in the genuine innovator product; B–F) surface of granules in the two generic samples with poor dissolution distributed in lower middle-income countries. Granules B and C were in a capsule from one sample; D, E, and F were in a capsule from another sample.
The actual physical properties of substandard and falsified medicines cannot be determined by only identifying differences from genuine products. We thought that if we could determine differences from the genuine product owing to the manufacturing process and further identify the raw materials used, we would be able to obtain clues for detecting substandard and falsified products. Therefore, we applied chemical imaging to visualize the component distribution and the internal structure of tablets. Thus, near-IR (NIR) imaging was used to visualize the ingredient distribution in tablets. In roxithromycin tablets with poor dissolution and poor disintegration,11) we found that although the content of the active ingredient was appropriate, some ingredients were aggregated and dispersed unevenly.26) We scanned the NIR spectrum of a flat cross section obtained by shaving the surface of roxithromycin tablets, and images were generated based on the first principal component score obtained using principal component analysis (PCA) of the standardized NIR spectrum data (Fig. 4). PCA was used because the excipients contained in tablets distributed in LMICs are not informed. In the innovator product, which had no problems in quality tests, the difference was small when comparing the NIR spectra of the red part with a high PCA score and the blue part with a low PCA score; this suggested no agglomeration of ingredients and uniform dispersion (Fig. 4A). However, in some generic medicines with poor dissolution that were distributed in one LMIC, large differences were observed in the NIR spectra of each part, indicating that the dispersion of ingredients was non-uniform and that there were parts where the components aggregated. The aggregates were confirmed by comparing the NIR spectrum of the part where aggregation was observed with the NIR spectra of the active pharmaceutical ingredient and the major excipients, revealing the existence of large aggregates of magnesium stearate and cornstarch (Fig. 4B). We also used Raman imaging to identify the aggregated components that were difficult to identify with NIR spectra.
An innovator product in which the components are finely dispersed (A), and a generic product in which the components are aggregated (B). The principal component analysis (PCA) score of the NIR spectrum obtained from each pixel is shown as the hue saturation value between red and blue. Higher PCA scores are redder in color, and lower PCA scores are bluer in color. The greater the color contrast, the clearer the bias in component dispersion.
The above findings highlighted the need to review the manufacturing process. These results also indicated the possibility of evaluating substandard and falsified medicines based on their formulation characteristics. Our research results yielded valuable information on the pharmaceutical situation in LMICs that can greatly contribute to the maintenance and improvement of pharmaceutical quality in these regions.
In research on the development of detection methods for substandard and falsified medicines, we focused on non-destructive analysis from the viewpoint of speed and simplicity, and we evaluated the applicability of NIR spectroscopy and Raman scattering analysis in the identification of falsified medicines. We examined tablets distributed in LMICs and on the Internet using several portable devices. Multivariate analysis of the obtained spectra clarified the possibility of this approach for identifying falsified medicines.21–24,27) The reason for using a portable device was to facilitate on-site inspection. Whereas evaluation with a high degree of accuracy using a large device is the mainstream approach, our research yielded a revolutionary approach of detection using a portable device. Later, with the development of an inexpensive, domestically produced, ultra-compact Raman spectroscopic module, we established a falsified medicine identification system using Raman scattering analysis and this device. The stability of data acquisition was improved by making a prototype focus guide and improving the device side.25) As samples, I used four kinds of erectile dysfunction therapeutic agent, Cialis, Viagra, Levitra and Diflucan. These were obtained by personal import via the Internet and included falsified medicines that were actually distributed. The obtained Raman spectra are shown in Fig. 5. In these spectra, some differences and obvious saturation could be observed, but it was difficult to visually evaluate their similarity and discriminate them all, even in spectra where sharp peaks appeared. I therefore tried using chemometric analysis. We examined discriminant analysis methods, considering the characteristics of existing falsified medicines. We investigated the applicability of two representative discriminant analysis methods, partial least squares discriminant analysis (PLS-DA) and soft independent modeling of class analogies (SIMCA). Using PLS-DA, if the prediction is successful, only a genuine sample will have a predicted value of 0.5 or more compared with the genuine model. In this analysis, falsified medicines and different products were falsely distinguished as genuine models (Fig. 6A). The reasons for this are thought to be because it is inappropriate to define falsified medicines from which various spectra are obtained in one category as well as owing to the small number of samples used in PLS-DA, which is supervised learning. However, in SIMCA using a PCA model (PCA-SIMCA), samples that fall within the limit of both the sample distance to the model (Si) and leverage (Hi), which shows the 95% confidence interval for the standard PCA model, are considered to belong to the model. In this analysis, all samples were correctly distinguished, revealing that PCA-SIMCA is more suitable for identifying falsified medicines28) (Fig. 6B). As a result, we established a discrimination algorithm for detecting falsified medicines, which is essential for social implementation, and developed a system of identifying falsified medicines using an ultra-compact Raman spectroscopic module. In the future, social implementation of this system will become even more realistic through building a spectrum library of genuine products and developing an application that executes SIMCA in real time.
Two Diflucan falsified samples were measured using both the front and back side of the tablet, and four spectral data were obtained.
A) Results of PLS-DA; a predicted value of 0.5 is shown as a red line, compared with the genuine Cialis model. B) Results of principal component analysis-soft independent modeling of class analogies (PCA-SIMCA); the 95% confidence interval for the standard PCA model is shown as a red line. The results of analysis using the genuine Cialis model are shown. Closed and open markers represent genuine and falsified samples, respectively. Squares, diamonds, triangles, and circles represent Cialis, Viagra, Levitra, and Diflucan, respectively.
Another approach is the development of small-scale quality test methods.29) Without the eradication of substandard and falsified medicines, it is impossible to solve the problem of access to medicines and achieve universal health coverage as stated in Sustainable Development Goal 3.8. However, LMICs with limited resources are unable to conduct pharmacopoeial tests. I attempted to develop a small-scale quality test method using a small number of samples and statistical techniques, as a quality testing method for pharmaceuticals that can be implemented in LMICs. Many substandard medicines that have been previously identified have poor dissolution, which is often found in falsified medicines. We therefore considered downsizing the dissolution test.
The first stage of the United States Pharmacopeia dissolution test requires a minimum of six tablets. We aimed to develop a screening method using three tablets. We first derived the criteria using metronidazole and validated these criteria using cimetidine. The medicines tested were collected in Cambodia. We sought to determine a reference value that could be used to absolutely judge acceptability or unacceptability of samples in the first stage.
As a first method, the lower limit of the 95% confidence interval for the average dissolution rate of three metronidazole tablets, which had a Q value of 85%, was used to examine the acceptance criteria. We derived an acceptance criterion based on a lower limit of the 95% confidence interval of 85% or more, that is, the Q value or more.
As a second method, we examined the criteria using the average dissolution rate of three tablets and the minimum dissolution rate. The criteria derived involved an average dissolution rate of 91% or more and a minimum dissolution rate of 87% or more; that is, the average dissolution rate is the Q value +6% or more and the minimum dissolution rate is the Q value +2% or more.
As a result of verification with cimetidine tablets, we concluded that the criteria for judging 100% of acceptable samples as being acceptable in the first stage and 100% of unacceptable samples as being unacceptable are that the average dissolution rate of three tablets is the Q value +6% or more and the minimum dissolution rate is Q value +2% or more.
In this small-scale quality test method, dissolution tests are performed using three tablets, against full-scale quality tests in accordance with the pharmacopoeia. This approach is positioned as a screening method for assessing drug quality. It can be expected to function as a test method for monitoring and guiding the quality control of pharmaceuticals in LMICs.
For social implementation of the falsified medicine identification system and small-scale quality test methods developed in these studies, it is necessary to expand the sample and evaluate the external validity. Currently, we are proceeding with the verification of medicines distributed in LMICs and medicines obtained via personal import from the Internet. Obtaining genuine products marketed in foreign countries is another challenge, and we are requesting the cooperation of manufacturers and distributors in our research.
We have also started to evaluate the applicability of novel methods for detecting substandard and falsified medicines using various analytic approaches that are not bound by conventional methods. Efforts to apply modern technology to the physical property evaluation of pharmaceuticals can contribute not only to finding new patterns of substandard and falsified medicines but also to expanding applications of this technology. At present, the basis for discriminating a falsified medicine is to distinguish it from the genuine medicine. If it is difficult to obtain genuine products, such as foreign products, the authenticity of substandard medicines cannot be confirmed even if they can be detected by referring to the pharmacopoeia. However, if there is no genuine product available, it is useful to request the manufacturer to authenticate their product. We have previously requested that manufacturers authenticate their products. Although we have been able to confirm the authenticity of some samples, the authenticity of other samples remains unknown.
In our investigation to assess the quality of medicines, we have found that the process of mixing excipients or the quality and selection of excipients themselves may be inappropriate for pharmaceutical products.26) To prioritize profit, falsified medicines might be composed of low-quality raw materials that are readily available rather than high-quality, expensive raw materials that are only suitable for manufacturing medicines. Additionally, falsified medicines are different from authentic ones in that management of the manufacturing process is insufficient, including in terms of hygiene, and there is a possibility of contamination with foreign substances in a poor manufacturing environment. It has become easier to obtain materials and equipment to manufacture falsified medicines, which are continually becoming more sophisticated. However, some characteristic or property always exists that cannot be imitated, and this can be used to distinguish medicines that are falsified from those that are genuine. Aiming to improve the differential accuracy, we are evaluating the applicability of various analytic methods and expanding the range of medicines to be verified. In this approach, detecting substandard medicines according to pharmaceutical characteristics that reveal differences in manufacturing processes could overcome the limits of detection methods that rely on comparisons with authentic products. In the future, I would like to clarify the limits of Raman scattering analysis, which is expected to be the most implementable approach, and develop a discrimination method that can compensate for those limits, thereby improving the accuracy of this method for discriminating substandard and falsified medicines.
It is the responsibility of pharmaceutical researchers to ensure the safety and security of pharmaceuticals. Preventing damage to health owing to substandard and falsified medicines and optimizing the distribution of pharmaceuticals are other issues that should be addressed by pharmaceutical researchers. Seven years have passed since Society 5.0 was proposed as a future society that Japan should aspire to in January 2016.31) The distribution of pharmaceuticals may change in the future with smart medical care. However, with the persistence and availability of substandard and falsified medicines and medicines that are improperly distributed without quality assurance, smart medical care will not be possible and the ideal future envisioned for society will not be realized. I believe achievement of the future medical care that society seeks is only possible by providing patients with high-quality pharmaceuticals to treat disease.
Through the research activities described above, we have demonstrated the possibility of detecting substandard and falsified medicines using various analytic methods that are not bound by conventional methods. Our efforts to apply modern technology to evaluation of the physical properties of pharmaceuticals can contribute not only to identifying newly available substandard and falsified medicines but also to expanding the application of these technologies. I will continue to contribute to the enhancement of security measures for medicines based on scientific evidence to help protect individuals from the risks of substandard and falsified medicines so that patients can receive the benefits of more effective and safe pharmaceutical treatment.
I would like to express my deep gratitude to Prof. Yoshimichi Sai (AI Hospital/Macro Signal Dynamics Research and Development Center, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University; Department of Hospital Pharmacy, Kanazawa University Hospital) for support and encouragement; Dr. Kazuko Kimura (previously professor of Drug Management and Policy, Faculty of Pharmacy, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University; Medicine Security Workshop) for lending her experience and expertise; Dr. Tsuyoshi Tanimoto (previously professor of Department of Analytical Chemistry, Faculty of Pharmaceutical Sciences, Department of Clinical Pharmacy, Doshisha Women’s College of Liberal Arts), Dr. Tatsuo Koide (Division of Drugs, National Institute of Health Sciences), Prof. Mikio Koyano (Japan Advanced Institute of Science and Technology), and Dr. Makoto Watanabe (Research Center for Structural Materials, Bonding and Manufacturing Field, National Institute for Materials Science) for technical assistance with the experiments and discussion; and Dr. Mohammad Sofiqur Rahman (Department of Clinical Pharmacy, School of Pharmacy, University of California, San Francisco), Dr. Shu Zhu (AI Hospital/Macro Signal Dynamics Research and Development Center, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University), Dr. Mirai Sakuda (Plus Pharmacy), and Dr. Tomoko Sanada (graduate of Drug Management and Policy, Kanazawa University) for collaboration from the early stages of this work. I am deeply grateful to my colleagues who cooperated in the field studies. This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Nos. JP26870220 and JP19KK0236, Shibuya Science Culture and Sports Foundation in FY 2019, and Heiwa Nakajima Foundation in FY 2019. Sample collection of Viagra, Cialis, Levitra, and Diflucan tablets was supported by Health and Labor Sciences Research Grants from the Ministry of Health, Labour and Welfare, Japan [Grant Nos. H23-chikyukibo-shitei-006 and H26-chikyukiboA-shitei-003]. The RXM tablets used in this study were collected with financial support from WHO Reference 2015/491209.0. The author thanks Analisa Avila for editing a draft of this manuscript.
The author declares no conflict of interest.
This review of the author’s work was written by the author upon receiving the 2023 Pharmaceutical Society of Japan Incentive Award for Women Scientists.
