<aside> 🔑 We're making novel leaps in target assessment and validation for oncology drugs to mitigate any unforeseen off-target effects early on in the drug development pipeline. By using machine learning to parse several papers and previous clinical trials, we can not only identify connections between different mechanisms of actions, cell lines, and off-target effects, but also propose key changes and recommendations for designing a trial that address other key problems such as patient screening by providing novel biomarkers for positive patient collection, repurposing a drug based on its off-target for novel usage, or simply learning more abut the drug prior to the trials.

</aside>

What are some of the more fundamental problems in clinical trials/why oncology?

• clinical trials take 6-7 years to complete, and while average FDA approval rate on average is 13.8%, oncology trials are at a disproportionately low level at 3.4%, with 97% of oncology trials failing in phase II-III

• when looking at the cases for why, oncology is one of the several growing therapeutic areas which require substantially larger amounts of funding at every stage, along with the fact that the growth and market cap of these riskier therapeutic areas are significantly larger (10.8 billion in 2019 with a CAGR of 5.4%), oncology makes up 35% of the entire clinical trial market size:

  

• fundamental problems that are prevalent in all clinical trials are heightened in oncology, such as the duration of the study as most drugs need to be evaluated over a longer-time period, the lack of patient accruement due to the stringent and unclear selection criteria (noted to be the problem for 60+% of all oncology trials found in clinicaltrials.gov), and most drug companies lack key information to make trials easier as most targets and mechanisms are often novel

Ultimately, the goal is to demonstrate efficacy and this is biologically more difficult for all current oncology trials and the drug development process as a whole.

• while there can be several major improvements in the logistical and technical parts of designing a clinical trial (Power of trial, statistical bias, funding, etc.), most are already being addressed yet are still far from being implemented and have found to be ineffective at solving some of the more fundamental problems with oncology trial success rates (lowest POS, second lowest is at 15% whereas the highest is just 27%)
• not only do we want to help oncology trials reduce costs by better designing trials, but also reap the benefits of phase III profits considering that oncology trials are a part of the ever-growing expanded-access trials

What are the key problems that we're trying to address?

Where can BenchSci win and how?