2021 年 3 巻 2 号 p. 30-35
The purpose of this paper is to visualize drug discovery research projects in academia in order to assess their progress and provide strategies for facilitating collaboration with industry. We divided academia’s drug discovery research data package into in vitro and in vivo milestones. Then, we set up the Drug Seeds Alliance Network Japan (DSANJ) original study data package required in each milestone by analyzing an oncology academic drug discovery research project. The Proof of Mechanism (PoM) data was the key to the realization of research alliances in Japan. We discuss ideal support systems to 1) visualize, 2) assess, and 3) revise the progress of academic research projects based on common milestones with industry partners.
We have been running the Drug Seeds Alliance Network Japan (DSANJ) Bio Conference since 2010 to support research collaboration between academia and pharmaceutical companies. The DSANJ Bio Conference (D-Bio) is a match-up conference that joins approximately 40 research-oriented pharmaceutical and venture companies. D-Bio has invited a total of 1,288 researchers, matched up 6,707 meetings, and achieved 65 research collaborations for 10 years. In this study, we summarize the progress of research projects in academia and discuss the key aspects for technology transfer to industry. We also discuss the benefits of constructing an academia support system through identifying common milestones.
In recent years, open innovation has become a major worldwide trend through expanding the source of drug seeds in pharmaceutical companies to academic drug discovery projects. The Japanese government established the Japan Agency for Medical Research and Development (AMED), and AMED has established a system to support and promote drug discovery projects in academia since 2015.A survey of new drugs launched in Japan over the past 40 years shows that the probability of launching compounds discovered through drug discovery research projects conducted independently by academia is significantly lower than that of compounds discovered through industry-academia collaborations [1].
Research and development (R&D) projects in pharmaceutical companies require large financial investments, and involve a considerable period of time as well as risk management before the product is launched. Pharmaceutical companies divide the R&D process into several stages and set milestones for each stage to strictly control the progress, including the decision to continue or abandon research projects. The classification of stages and milestones is specified in detail by each pharmaceutical company, but these are not disclosed to the public. Based on this background, we theorized that if there were some common milestones for both parties, this would promote industry-academia collaborations, and we published the basic R&D stages and milestones with the cooperation of several pharmaceutical companies in 2018 [2]. A summary of 10 basic drug discovery and development R&D stages and milestones is presented in Table 1. These guidelines may be applied to all diseases; however, the study data that needs to be acquired varies according to the research topic. For example, in the case of oncology therapeutic agents, study data showing the effect of suppressing the growth of cancer cells at each milestone are required; in the case of inflammatory disease therapeutic agents, study data showing the effect of suppressing substances that cause inflammation at each milestone are required.
| Stage | Milestones to be achieved (major categories) | |
|---|---|---|
| 1 | A | Formulation of a research plan from the conception of a drug target |
| B | Discovery and identification of drug targets | |
| C | In vitro validation of drug targets | |
| D | In vivo validation of drug targets | |
| 2 | A | Assay system construction |
| B | Construction of an HTS screening system (confirmation of z-value) | |
| C | Screening of compounds (source/type of library, number of compounds) | |
| D | Selection of hit compounds (compounds considered promising based on activity and chemical structure) | |
| 3 | A | Expand to seed compounds (compounds in which the same type of activity is found in a series of derivatives) |
| 4 | A | Development from seed compounds to lead compounds (compounds whose activity, selectivity, toxicity, and physical properties can be improved by structural transformation) |
| 5 | A | Discover and optimize lead compounds that show efficacy in animal studies |
| 6 | A | Identifying and optimizing several promising compounds to proceed to non-GLP studies |
| 7 | A | Conduct non-GLP studies (pharmacodynamics, ADMETox, physical properties) and select P0 candidate compounds |
| B | Establishment of a mass synthesis method (production method) | |
| 8 | A | Conduct GLP studies (pharmacodynamics, ADMETox, physical properties, etc.) and select P1 candidate compounds |
| B | Establishment of a GLP mass synthesis method (production method) | |
| C | Formulation study | |
| 9 | A | PoM acquisition of compounds |
| B | Determination of dosage formulation | |
| 10 | A | PoC acquisition of compounds |
HTS, high throughput screening; ADMETox, absorption, distribution, metabolism, excretion and toxicology; GLP, good laboratory practice; PoC, proof of concept.
Therefore, we hypothesized that a study data list with milestones required for each disease would provide a common theme that facilitates industry-academia collaborations. If this study list is updated in real-time, it could allow for simple and quick visualization and assessment of the progress of drug discovery R&D projects. This approach improves research efficiency by cost-saving and reducing the total research time. The summary presented in this work is a culmination of 10 years of work analyzing 65 research collaborations from the DSANJ [3], and a specific example was selected for this study in the field of oncology research.
The progress of the drug discovery research project was visualized using the original study data package that is used internally by DSANJ. We focused on the in vitro and in vivo stages (Table 1, stages 1-C and 1-D, respectively) because our track record at DSANJ indicates that these two phases of research are the most important in creating meaningful collaborations between academia and pharmaceutical companies [4, 5]. We selected the oncology drug discovery research topic as a specific example since there are many research projects in this field in Japanese academia. Analysis of the academic research project involved dividing the milestones of the drug discovery process into in vitro and in vivo sections. The DSANJ original study data package was applied for each milestone and we refer to the Koyanagi’s group report in the AMED report [2]. We verified the validity of the DSANJ original study data package required in each segmented milestone using a successful research collaboration case at DSANJ.
DSANJ original study data package required in vitro validationTable 2 shows the list of DSANJ original study data packages required in each segmented milestone for in vitro validation of drug targets in oncology research, with six milestones required to complete this study data package.
| Milestone (Segmented) | Study data package (Required) |
|---|---|
| 1. Proof of Mechanism (PoM) Validation of drug targets by using research tools (small molecule compounds, siRNA, antisense, antibodies, proteins, peptides) | • Checking the validity of research tools used for verification. |
| • Inhibition of proliferation in cancer cells. | |
| • Confirmation of inhibitory effects on the functions of peri-cancer cells (cancer stem cells, cancer-associated fibroblasts, etc.) that affect cancer growth. | |
| • Confirmation of the effects on normal cell proliferation. | |
| • Confirmation of the effects on normal cell function. | |
| 2. Collection of information on species differences and examination of species differences (investigation of extrapolation to humans through gene sequence/amino acid sequence analysis) | • Examination of species differences (evaluated in a human cancer cell transplantation model). |
| • Consideration of species differences (when the target is peri-cancer cells or when in vivo drug efficacy evaluation in animal models of spontaneous cancer is required). | |
| • Collection of information on species differences through literature and databases. | |
| 3. Comparison with existing drugs (existing standard drugs or reagents) Case 1: When the mechanism of action is the same as that of existing drugs Case 2: When the mechanism of action is different from that of existing drugs | <Case1> |
| • Drug target evaluation. | |
| • Inhibition of proliferation in cancer cells. | |
| • Confirmation of inhibitory effects on the functions of peri-cancer cells (cancer stem cells, cancer-associated fibroblasts, etc.) that affect cancer growth. | |
| • Confirmation of the effects on normal cell proliferation. | |
| • Confirmation of the effects on normal cell function. | |
| <Case2> | |
| • Confirmation of the action of existing drugs on the targets. | |
| • Inhibition of proliferation in cancer cells. | |
| • Confirmation of inhibitory effects on the functions of peri-cancer cells (cancer stem cells, cancer-associated fibroblasts, etc.) that affect cancer growth. | |
| 4. Confirmation of the effectiveness of combining with existing drugs | • Confirmation of the effectiveness of combining with existing drugs. |
| 5. Verification of the action leading to the drug development of the target in question using patient-derived cells (tissue) | • Confirmation of target protein/gene expression in patient-derived cells (tissues). |
| • Functional verification of target proteins/genes in patient-derived cells (tissues). | |
| 6. Data or literature information that clearly shows a relationship to the disease state (including the possibility of serious side effects) | • Clinical data/epidemiological study results conducted by a PI. |
| • Information from literature and databases. |
Table 3 shows the list of DSANJ original study data packages required in each segmented milestone for in vivo validation of drug targets in oncology research, with four milestones required to complete this study data package.
| Milestone (Segmented) | Study data package (Required) |
|---|---|
| 1. In vivo validation of drug targets by using research tools (small molecule compounds, siRNA, antisense, antibodies, proteins, peptides) | • Confirmation in vivo PK-PD of research tools used for validation. |
| • Inhibition of tumor growth in a mouse xenograft model. | |
| • Inhibition of tumor growth in animals with spontaneous cancer. | |
| • Checking for effects on normal animals. | |
| 2. Preparation of animal models for cancer | • Xenograft model for transplantation of novel human cancer cells. |
| • Xenograft model for transplantation of human organelle containing cancer cells. | |
| • Genetically engineered cancer models. | |
| 3. Comparison with existing drugs (existing standard drugs or reagents) | • Confirmation of in vivo PK/PD of existing drugs used in comparative studies. |
| • Comparison of tumor growth inhibition in a mouse xenograft model. | |
| • Comparison of tumor growth inhibition in animals with spontaneous cancer. | |
| • Check for effects on normal animals. | |
| 4. Confirmation of the effectiveness of combining with existing drugs | • Confirmation of the effectiveness of combining with existing drugs. |
| • Checking for effects on normal animals. |
PK-PD, pharmacokinetics and pharamcodynamics.
This case of an academic research project was invited to DSANJ twice. In the first invitation, several pharmaceutical companies showed a high level of interest but did not collaborate. The second invitation, in which the researcher in academia obtained additional data, resulted in a research collaboration.
The study data in this research project were all matched with the DSANJ original study data package (Tables 2 and 3), as shown in Table 4. This data was used to analyze an example of an academic research project that led to a research alliance at DSANJ. At the first invitation to the DSANJ Bio Conference, there was no in vivo validation data. Instead, in vitro proof of mechanism (PoM) validation showed the specificity of the antibodies used against the target protein and the inhibitory effect of the antibodies on cancer cell growth (Table 4, yellow highlighted text). In addition, it was shown that a glycoprotein is expressed in cancer cells derived from patients.
| Milestone (Segmented) | Study data package (Required) |
|---|---|
| In vitro (stage 1-C) | |
| 1. Proof of Mechanism (PoM) validation of drug targets by using research tools (small molecule compounds, siRNA, antisense, antibodies, proteins, peptides) | • Checking the validity of research tools used for verification. |
| • Inhibition of proliferation in cancer cells. | |
| • Confirmation of inhibitory effects on the functions of peri-cancer cells (cancer stem cells, cancer-associated fibroblasts, etc.) that affect cancer growth. | |
| • Confirmation of the effects on normal cell proliferation. | |
| • Confirmation of the effects on normal cell function. | |
| 2. Collection of information on species differences and examination of species differences (investigation of extrapolation to humans through gene sequence/amino acid sequence analysis) | • Examination of species differences (evaluated in a human cancer cell transplantation model). |
| • Consideration of species differences (when the target is peri-cancer or when in vivo drug efficacy evaluation in animal models of spontaneous cancer is required). | |
| • Collection of information on species differences through literature and database searches. | |
| 3. Comparison with existing drugs (existing standard drugs or reagents) Case 1: When the mechanism of action is the same as that of existing drugs. Case 2: When the mechanism of action is different from that of existing drugs | <Case1> |
| • Targeting considerations. | |
| • Inhibition of proliferation in cancer cells. | |
| • Confirmation of inhibitory effects on the functions of peri-cancer cells (cancer stem cells, cancer-associated fibroblasts, etc.) that affect cancer growth. | |
| • Confirmation of the effects on normal cell proliferation. | |
| • Confirmation of the effects on normal cell function. | |
| <Case2> | |
| • Confirmation of the action of existing drugs on their targets. | |
| • Inhibition of proliferation in cancer cells. | |
| • Confirmation of inhibitory effects on the functions of peri-cancer cells (cancer stem cells, cancer-associated fibroblasts, etc.) that affect cancer growth. | |
| 4. Confirmation of the effectiveness of combining with existing drugs | • Confirmation of the effectiveness of combining with existing drugs. |
| 5. Verification of the action leading to the drug development of the target in question using patient-derived cells (tissue) | • Confirmation of target proteins/gene expression in patient-derived cells (tissues). |
| • Functional verification of target proteins/genes in patient-derived cells (tissues). | |
| 6. Data or literature information that clearly shows a relationship to the disease state (including the possibility of serious side effects) | • Clinical data/epidemiological study results conducted by a Principal Investigator (PI). |
| • Information from literature and databases. | |
| In vivo (stage 1-D) | |
| 1. In vivo validation of drug targets by using research tools (small molecule compounds, siRNA, antisense, antibodies, proteins, peptides) | • Confirmation in vivo PK/PD of research tools used for validation. |
| • Inhibition of tumor growth in a mouse xenograft model. | |
| • Inhibition of tumor growth in animals with spontaneous cancer. | |
| • Checking for effects on normal animals. | |
| 2. Preparation of animal models for cancer | • Xenograft model for transplantation of novel human cancer cells. |
| • Xenograft model for transplantation of human organelle containing cancer cells. | |
| • Genetically engineered cancer models. | |
| 3. Comparison with existing drugs (existing standard drugs or reagents) | • Confirmation of in vivo PK/PD of existing drugs used in comparative studies. |
| • Comparison of tumor growth inhibition in a mouse xenograft model. | |
| • Comparison of tumor growth inhibition in spontaneous cancer animals. | |
| • Check for effects on normal animals. | |
| 4. Confirmation of effectiveness of combining with existing drugs | • Confirmation of effectiveness of combining with existing drugs. |
| • Checking for effects on normal animals. |
At the second invitation to the DSANJ Bio Conference about two years later, the data showing “Antitumor effects in an in vivo breast cancer transplant model” were added to the proposed material (Table 4, blue highlighted text). In other words, the in vivo PoM validation was confirmed, suggesting that expressing the study data acquired in academia in the DSANJ original study data package is useful for visualizing the progress of drug discovery research projects in academia. This specific oncology research example showed that only some of the study data in each segmented milestone were acquired, indicating that acquiring both in vitro and in vivo PoM study data are key to research alliances. The remaining study data packages listed in Tables 2 and 3 are useful for some pharmaceutical companies to evaluate drug discovery research projects in academia.
The DSANJ original study data package and segmented milestones may be defined as “common milestones” for technology transfer. Terashita [6] stated that there are many cases where interesting drug discovery research projects do not proceed to research collaboration due to a lack of key corporate standard study data. Therefore, it is important for both parties to evaluate drug discovery research projects by the same standards in order to transfer academia’s potential drug seeds to pharmaceutical companies. We propose that the common milestones may serve as a basis for successful technology transfer between academia and pharmaceutical companies. Next, we discuss a novel supporting forms to facilitate successful drug discovery research projects in Japan academia.
Proposal I: A novel support system to visualize the progress of research projects based on common milestonesThe first step for a meaningful academic-industry collaboration involves determining a list of common milestones so that academics may conduct experimental research according to the plan. Support organizations such as AMED could assist researchers in academia by providing a way of visualizing the progress of the project in real-time based on the common milestones. This is important because in Japan, there are more than 90 universities, colleges, and institutions involved in promoting drug discovery research projects, all of which are located in disparate places, and the number of human resources with expertise in drug discovery in Japan is currently limited. We propose that securing specialized human resources and dispatching the relevant experts for the particular drug discovery project could lead to the optimal allocation of drug discovery resources throughout Japan within the limited national budget.
Proposal II: A novel support system to accurately assess research project achievementsThe progress of the research project needs to be accurately assessed in the context of addressing the common milestones through discussions between researchers and pharmaceutical company representatives. Experts in drug discovery in support organizations such as AMED could assist researchers in comprehensively understanding any issues or requirements raised by pharmaceutical companies. Video-communication services such as Zoom, Cisco Webex, Microsoft Teams, and Google Meet are now widely available, which could foster meaningful discussions between academics and pharmaceutical companies and lead to improved productivity.
Proposal III: A novel support system to revise research plans and promote research alliancesAfter accurately assessing the progress of their research project, researchers in academia will be able to revise their research plans and pursue more productive and efficient research. The efforts in the cycle of (1) visualizing the progress of research projects based on common milestones, (2) accurately assessing the degree of progress through discussion with pharmaceutical companies, and (3) revising the research plan could be seen as a concrete method to make effective use of research time and budget.
In conclusion, we propose that following the virtuous cycles of (1) visualization, (2) assessment, and (3) revision based on common milestones would be a novel support system for academic researchers and support organizations such as AMED for accelerating research alliances within industry.
The authors have nothing to disclose.
This research was supported by AMED under the Grant Number JP20nk0101405. We would also like to express our gratitude to the Japan Pharmaceutical Manufacturing Association (JPMA), Osaka Chamber of Commerce and Industry (OCCI), and AMED for their dedicated contribution through the DSANJ Bio Conference to promote research alliances in Japan.
