Discovering new drugs, developing them, and bringing them to the market is a five-step process. Each step enables crucial functions to be performed that ensure that new drugs are safe and effective.

Step 1: Discovery and Development

The discovery phase for new drugs is a fascinating one that incorporates many different insights into disease processes and potential treatments. According to the U.S. Food & Drug Administration, during the discovery and development phase, thousands of compounds can have potential promise for the treatment of various diseases. However, as the discovery phase continues, only a small number of compounds usually warrant additional study.

Finding the compounds that show the highest level of promise for ongoing research is a challenging process due to sparse data and uncertainty. This is where artificial intelligence (AI) can assist. It can help to uncover hidden opportunities caused by missing, uncertain, or inaccurate data and prioritize the most valuable experiments. Among AI-augmented drug development software for drug discovery, Cerella uses a unique deep learning approach that can help to overcome limitations in drug discovery data.

Once the most promising compounds are identified through efficient and effective analysis of data gathered during the discovery process, the development phase continues. The compounds identified during drug discovery are then tested for how they meet the following criteria:

• Absorption, distribution, and metabolization
• Potential benefits and mechanism of action
• Dosage levels
• Optimal administration methods
• Potential adverse effects
• Effects on different groups of people
• Interactions with other drugs or treatments
• Effectiveness when compared to drugs of a similar class or type

Step 2: Preclinical Research

Following the development phase, preclinical research focuses on whether a potential drug could cause harm. Preclinical research is conducted both in vitro (Latin for “in glass” – meaning in the laboratory) and in vivo, in living organisms.

U.S. Food & Drug Administration regulations for preclinical research operate under good laboratory practices (GLP) that are set forth in 21 CFR Part 58.1: Good Laboratory Practice for Nonclinical Laboratory Studies. GLP regulations cover study conduct, personnel, facilities, equipment, written protocols, operating procedures, and study reporting procedures. They also specify quality assurance systems for preclinical research on FDA-regulated products.

At the end of preclinical research, researchers review the findings and decide if the drug should be advanced for further research.

Step 3: Clinical Research

After the conclusion of in vitro and in vivo preclinical research, a drug that shows enough promise to advance can enter the clinical research phase. This phase is the part of the drug discovery process that the public is probably most familiar with. It includes clinical drug trials that are done on people. A drug developer will initiate an Investigational New Drug Application (IND) to the Food & Drug Administration prior to clinical research studies.

Clinical trial design covers who will participate, how many will participate, and how long the study will last. There are three phases of a clinical trial, from Phase 1 (early, small scale) to Phase 3 (later stage, large scale). About 70% of drugs move from the earliest phase onto Phase 2.

Phase 2 clinical research studies incorporate a larger group of people that could potentially benefit from a new drug. They can last up to two years, and only about a third of Phase 2 drugs move on to Phase 3. In Phase 3 of clinical research, several thousand volunteers can participate, and monitoring can last up to four years. Within this stage, new technologies are also becoming more prevalent, such as wearable technology like smartwatches that can reliably collect data with minimal input from the participant. For example, an Apple Watch heart study by Stanford medicine researchers could safely identify heart rate irregularities and use this data in their clinical evaluations.

At the end of this phase, between 25% and 30% move on to further testing or are deemed safe and effective enough to warrant a marketing application. At least two large-scale clinical research studies are required to advance a drug to the next step in the process.

Step 4: FDA Review

The FDA has undertaken steps to streamline its approval process for new drugs, but following the completion of clinical research, the New Drug Application (NDA) is the next step. The NDA is supposed to tell the “full story” of the drug. It should include reports on all previous studies, including preclinical data. The NDA should also include proposed labeling, drug abuse information, patents, and IRB (Institutional Review Board) compliance reports.

The FDA’s review team will approve the application and can take up to six to ten months to respond. The review team can request additional studies, and they can also ask the drug developer to explain the data. The review team also works with the drug developer to create labeling and prescribing information.

If the drug is approved, it can be marketed, but the FDA’s oversight doesn’t stop there.

Step 5: FDA Post-Market Safety Monitoring

After a drug is approved by the FDA, manufacturers need to file supplemental applications for any changes in the original NDA (New Drug Application). This can include changes in formulation or strength. One example would be a drug that is sold in different forms, for example, as an oral liquid or as an injectable. The FDA also conducts inspections of drug manufacturing facilities in the U.S. and in other nations for products sold in the U.S. When the patent of various drugs expires, this will enable other manufacturers to develop a generic version of the drug. Generic manufacturers don’t need to follow the entire drug development process, but they do need to conduct bio-equivalent studies, and they also need to submit a New Drug Application (NDA) to the FDA for approval of their generic drug product.

At the Post-Market Safety stage, the FDA is also developing new initiatives like the Sentinel Initiative, which uses large health records databases to monitor drug use and safety in real-time.