• There is very high-correlation between high patent count, clinical trials, increase in survival rates and reduction in absolute deaths for most cancers
• There is also a very high co-relation between high patent count and NIH funding of research
Methodology and data sources:
We have collated data from the following sources:
a) Clinical trial data from clinicaltrials.gov (56000+ records) and classified them by focus on cancer types (using MESH terms)
b) Patents data as provided by USPTO classified by cancer types, the genes they were focusing on by mapping to MESH/Wikipedia thesauri (mapped onto Title, Abstract, Claims and Example for high precision)
c) Classified the NIH funded projects by cancer type (Step (b)
d) Survival data and Absolute death data by cancer type from SEER was plotted against number of patents and clinical trials (classified by cancer type)
e) Overall Incidence rate and Incidence rates in working age population from CDC cancer registry (by cancer type) were plotted against number of patents and clinical trials
f) Mapped patents classified by cancer type to genes co-related to specific cancers (as defined in KEGG cancer pathway database)
g) Generated key concepts associated with USPTO provided patent list. Used expert input to filter tags for treatment and diagnosis methods and generated clusters for insights
h) Using data from Wikipedia/other sources, we classified the 1090 patents from USPTO data mapped to FDA orange book drugs into those approved for cancer or other disease types (852 patents mapped to 360 drugs related to cancer)
Data interpretation and insights
Types of Cancer Against Research and Funding Activity
We have used Bubble charts as a framework for slicing and dicing data. The 3 dimensions used in Bubble charts are:
a) Number of patents (classified by cancer type)
b) Funding (classified by cancer type)
c) Survival rate or Incidence rate or Absolute deaths or Clinical trials
Insights relating patents to survival, absolute deaths, incidence and clinical trials
• High patent activity leads to high survival and low absolute death. High patent counts and incidence rates are not correlated, perhaps as most patents deal with cancer cure than prevention.
• We find high patent counts co-relate strongly with high number of clinical trials.
Example insights for research priority:
a) Lung cancer (funding above median) has low survival rate of 17%, has the highest absolute death rate of 46 per 100,000 and a relatively high incidence rate (overall and in working age)
b) Colorectal cancer (funding below median) while has a median survival rate of 65% has a high death rate of 15 and a high incidence rate 41 (overall and in working age)
c) Breast cancer (high funding) has a very high incidence rate of 67, a high survival rate of close to 90% and yet a high mortality of 88. High patent activity, high clinical trials has lead to high survival rates
From above, can firmly conclude that Lung Neoplasm should be a clear priority for research. It is followed by Pancreatic Neoplasm and by Colorectal cancer (with a 65% survival rate) as areas to focus on. Better funding can lead to better outcomes.
Insights from approved drugs, success, policy guidance
• Out of 270,000 odd patents given by USPTO, only 1090 patents had an associated approved drug from FDA
• Out of these 1090 patents, 852 patents were associated with 360 approved drugs/combination drugs for cancer treatment. Of the 360, only 16 were from NIH funded projects.
• The NIH funding totaled 42 billion dollars (from NIH reporter)
• Most of this funding (close to 97% projects) was for University, Hospitals or research centers. Private firms received very little funding despite an overall high success rate (high drug approval rates)
• Perhaps, NIH could divert more of their research funding to private enterprise (large corporations, small firms and inventors)
• Off the 852 patents (a significant number as seen from re-assignment history) were re-assigned from a small firm/inventors to a large company, indicating that small companies are more innovative with early stage research while large companies are experts at managing the regulatory landscape (FDA approval process, funding late stage clinical trials etc.).
• NIH could thus focus on funding research across smaller firms
Insights from time lag between patents to products, patents to clinical trials
• The mean time taken from application date of a patent to FDA approval of the drug is 5 years, 8 months and 23 days
• The median time for the same is 4 years, 11 months and 18 days
• As indicated, most of the early and successful research is done by small to medium sized firms with limited resources to hire the best legal advice for patent prosecution. Any support to small firms to expedite the patent prosecution by USPTO may ease the administrative burden that small companies are least geared for.
• High patent activity is strongly co-related with high clinical trials
• High clinical trials are strongly co-related with drug approvals
Insights from breakthrough technology research and funding
• There is a very high co-relation between NIH funding and patent counts of treatment technology for instance immunotherapy has the highest patent count and also has the highest funding, followed by stem cell therapy and so on in the decreasing order of funding
• Immunotherapy and gene therapy/editing are very popular
• CRISPR/CAS9 gene editing is emerging as a promising sub-area under gene-editing
• Treatment devices for Hyperthermia, Cyrosurgery, Lasers and other methods are in relative infancy and can do with better funding
• High-throughput assays are leading diagnostics patent count and is also the most well funded
• Predictive diagnosis for predicting onset of cancer prior to its on-set should see a spike in patent activity given high funding
Insights from mapping patent research to ‘pathway data’/Genes research
• Co-relating patent research activity for a particular cancer type to pathway data gives hints of under-explored areas for research
• For instance, for lung cancer most patent research is focused on P53 gene (which is also popular for other cancer types). Lung cancer research on other genes can be encouraged.