Oncology Drugs Have Lower Approval Success Rates than Other Conditions

Study of clinical development success rates finds phase 2 trials are less successful than any other R&D phase.

Phase 2 drug trials were found to be less successful than any other phase of drug development, a recent study found.

Biotechnology Innovation Organization (BIO) released new findings from the largest study on clinical drug development success rates that aimed to strengthen benchmarking metrics for drug development.

Between 2006 and 2015, researchers recorded 9985 clinical and regulatory phase transitions from 7455 development programs across 1102 companies within the BioMedTracker database.

The BioMedTracker monitors the clinical development and regulatory history of investigational drugs to assess the likelihood of approval (LOA) by the FDA.

The diseases included in the report included allergy, autoimmune, cardiovascular, chronic high prevalence diseases, endocrine, gastroenterology, hematology, infectious disease, metabolic, neurology, oncology, ophthalmology, psychiatry, rare diseases, respiratory, and urology.

The study found that phase 2 success rates were significantly lower (30.7%, n=3862) than any other phase, a finding that consistent with previous studies. Phase 1 and 3 success rates were found to be higher than phase 2, with phase 1 slightly higher than phase 3.

Out of the 4 developmental phases, the highest success rate was found in the NDA/BLA filing phase (85.3%, n=1050).

One key finding of the study was that phase 3 trials have the second lowest transition success rates into an NDA/BLA filing. Researchers noted the significance, because most company-sponsored phase 3 trials are the longest and most expensive, accounting for 60% of all clinical trial costs.

Additionally, 35% of all R&D spending is currently on phase 3 development. Phase 1 trials were found to have just a 1 in 10 (9.6%) chance of advancing all the way to FDA approval. Researchers also separated major disease areas and categorized 21 major diseases and 558 indications between 2006 and 2015.

Only major diseases with more than 100 total transitions from phase 1 to NDA/BLA approval were analyzed, resulting in 14 disease categories: allergy, autoimmune, cardiovascular, endocrine, hematology, infectious disease, gastroenterology (non-IBD), metabolic, neurology, oncology, ophthalmology, psychiatry, respiratory, and urology.

Diseases with less than 100 total transitions, such as dermatology, dental, obstetrics, orthopedics, renal, and rheumatology (for non-autoimmune indications) were put into the “other” category.

The LOA from phase 1 was found highest in the hematology group (26.1%, n=283). A significant portion of hematology transitions came from anemia, blood protein deficiencies, hemophilia, hemostasis, and thrombocytopenia. The success rate of hematology LOA from phase 1 was 5 times the success rate of oncology, which had the lowest of all the major diseases.

There were several hemophilia indications with overall LOA that was more than 50%, offsetting the weaker hematology success rates found in venous thromboembolism and neutropenia.

The second highest LOA was infectious disease with 19.1% (n=916). The 4 lowest overall LOA that was under 9.6% included neurology, cardiovascular, psychiatry, and oncology, respectively.

Researchers note that both neurology and oncology have the highest n values, but also low LOA values, suggesting that these 2 disease categories play a significant role in bringing down the overall industry of LOA.

Success rates for phase 1 ranged from 53.9 to 84.8% with an average of 63.2%. Phase 2, which had the lowest transition success rates, ranged from 23.7 to 56.6%.

The success rates for NDA/BLA ranged from 77.5% for ophthalmology to 94.6% for respiratory. The 17.1% distribution of rates was the smallest among the 4 phases.

Researchers also examined how many original NDA/BLA filings were approved on first review. Psychiatry had a 37% chance of being approved on the first cycle, compared with an approximately 80% chance in oncology. Both categories resulted in 91% of original drug indication applications being approved.

The time from filing varied among the disease categories, with neurology drugs taking the longest to be approved at an average of 2 years, while oncology was approved almost twice as fast at 1.1 years. Since oncology had an outsized effect on the overall industry success rate, researchers compared phase transition success rates and LOA for non-oncology development programs against oncology development programs alone.

The cause of oncology having the lowest overall success rates out of all 14 disease categories was because of phase 3 transition success rates, which was 23% lower in the non-oncology disease areas, according to the study. Researchers noted a pullback in funding from investors for chronic high prevalence diseases, so they examined whether this issue was occurring because of the difficulty in the development and approval of medications to treat these diseases.

Chronic high prevalence diseases had low LOAs of 8.7% and contributed in part to the low 9.6% overall industry LOA. The transition success rates were found to be lower in phase 1 (58.7% vs 63.2%) and phase 2 (27.7% vs 30.7%) versus the overall dataset.

However, the rates were slightly higher in phase 3 and NDA/BLA with 61.6% vs. 58.1% for phase 3 and 87.2% vs. 85.3% for NDA/BLA. Chronic high prevalence diseases have higher success rates in phase 3 transition compared with all the other diseases, but lower than non-oncology transition success rates.

Additionally, the 11.9% LOA for all non-oncology indications was 3% higher than chronic high prevalence diseases. The findings suggest that chronic indications negatively impact overall success rates outside of oncology.

The use of biomarkers as inclusion or exclusion criteria (selection biomarkers) increased dramatically in enrolling patients into clinical studies. The benefit from selection biomarkers showed that the 1 in 4 LOA in phase 1 increased compared with 1 in 10 when no selection biomarker was used.

Researchers annotated the drugs in the dataset by new molecular entity (NME), non-NME, and biologic and vaccine. The probability of success between NME, biologic, and non-NME revealed that NMEs had the lowest transition success rates in every phase of development and had a 6.3% LOA from phase 1, while biologics had an LOA of 11.6%.

Non-NMEs had the highest LOA from phase 1 at 22.6%, with transition success rates well above the NME and biologic classifications in all phases. Researchers noted that many non-NMEs start development in phase 2 or 3, so the approval rate is most likely higher. Vaccines had a 16.1% LOA from phase 1.

Some of the potential factors that could impact the success rates in the report include patient population and selection strategy; clinical validation of a target, drug class, or mechanism of action; clinical trial complexity; lack of funding; and commercial decisions and portfolio prioritization, the study concluded.