`
`$33571”
`Organization
`
`1t.
`Biomedtrackerfi- ‘-
`
`9 AM PLION
`
`.
`
`.55;
`JP—
`
`‘J-_ Clinical Development
`Success Rates
`2006-2015
`
`mm mm Won-"a
`
`Genentech 2109
`Celltrion v. Genentech
`IPR2017-01122
`
`
`
`About BIO
`
`BIO is the world’s largest trade association representing biotechnology companies, academic institutions, state biotechnology
`centers and related organizations across the United States and in more than 30 other nations. BIO members are involved in
`
`the research and development of innovative healthcare, agricultural, industrial and environmental biotechnology products. BIO
`also produces the BIO International Convention, the world’s largest gathering of the biotechnology industry, along with industry-
`Ieading investor and partnering meetings held around the world.
`
`About Biomedtracker
`
`BioMedTracker, a subscription-based product of lnforma, tracks the clinical development and regulatory history of investigational
`drugs to assess its Likelihood of Approval (LOA) by the FDA. BioMedTracker is populated in near real-time with updated
`information from press releases, corporate earnings calls, investor and medical meetings and numerous other sources.
`
`About Amplion
`
`Amplion is the leading biomarker business intelligence company, and its flagship product BiomarkerBase“, along
`with consulting services and free reports, deliver insights that inform key strategic decisions for drug and diagnostic
`test developers. Since 2012 Amplion has helped large and small companies alike make the best use of biomarkers in
`advancing precision therapeutics and next generation diagnostics. BiomarkerBase is a subscription-based service that
`tracks biomarker usage in clinical trials, drug labels, and tests (including laboratory-developed, F DA—cleared, and FDA-
`approved tests). BiomarkerBase is updated weekly with information from these sources and publications, using supervised
`machine learning algorithms for natural language processing (Amplion BiomarkerEngine) to identify biomarkers.
`
`
`
`Executive Summary
`
`This is the largest study of clinical drug development success rates to date. Over the last decade, 2006-2015, a total of 9,985
`clinical and regulatory phase transitions were recorded and analyzed from 7,455 development programs, across 1,103 companies
`in the Biomedtracker database. Phase transitions occur when a drug candidate advances into the next phase of development or is
`suspended by the sponsor. By calculating the number of programs progressing to the next phase vs. the total number progressing
`and suspended, we assessed the success rate at each ofthe four phases of development: Phase I, II, III, and regulatory filing.
`Having phase-by-phase data in hand, we then compared groups of diseases, drug modalities and other attributes to generate
`the most comprehensive analysis yet of biopharmaceutioal R&D success.
`
`This work was made possible due to the years of clinical program monitoring and data entry by Informa’s Biomedtracker service.
`BlO has long partnered with Biomedtracker to calculate success rates based on this data. More recently, BIO and Biomedtracker
`partnered with Amplion, the inventors of BiomarkerBase, to analyze the effects of biomarkers in clinical trial success.
`
`Key takeaways:
`
`0
`
`The overall likelihood of approval (LOA) from Phase I for all developmental candidates was 9.6%, and 11.9% for
`all indications outside of Oncology.
`
`0 Rare disease programs and programs that utilized selection biomarkers had higher success rates at each
`phase of development vs. the overall dataset.
`
`0
`
`Chronic diseases with high populations had lower LOA from Phase lvs. the overall dataset.
`
`0 Of the 14 major disease areas, Hematology had the highest LOA from Phase I (26.1%) and Oncology had the
`lowest (5.1%).
`
`o
`
`Sub-indication analysis within Oncology revealed hematological cancers had 2x higher LOA from Phase I
`than solid tumors.
`
`0 Oncology drugs had a 2x higher rate of first cycle approval than Psychiatric drugs, which had the lowest percent
`of first-cycle review approvals. Oncology drugs were also approved the fastest of all 14 disease areas.
`
`0
`
`Phase ll clinical programs continue to experience the lowest success rate of the four development phases, with
`only 30.7% of developmental candidates advancing to Phase III.
`
`I Biotechnology
`Innovation
`Organization
`
`Biomedtracker
`Pharmo intelligence 1 informa
`
`GAMPLION
`
`BiUinCin-
`
`./
`
`-1:vralvsis
`
`i 3
`
`
`
`
`Bisease areas covered in this report:
`
`.ur'J
`
`- Allergy
`
`- Autoimmune
`
`. Cardiovascular
`
`- Chronic High Prevalence Diseases
`
`. Endocrine
`
`- Gastroenterology
`
`- Hematology
`
`.
`
`Infectious Disease
`
`. Metabolic
`
`. Neurology
`
`. Oncology
`
`- Ophthalmology
`
`- Psychiatry
`
`. Rare Diseases
`
`r .A 1.
`
`. Respiratory
`
`- Urology
`
`
`
`
`
`Table of Contents
`
`Introduction
`
`6
`
`Phase Success and Likelihood of Approval (LOA) —Overa||7
`
`Phase Success and Likelihood of Approval (LOA) — byDisease 8
`
`Oncology and Non-Oncology Diseases13
`
`Rare and Chronic High PrevalenceDisease16
`
`Patient Selection Biomarkers.............................................................................................................. 18
`
`Phase Success and Likelihood of Approval (LOA) - by Drug Classification
`
`20
`
`Discuss10n22
`
`Methods 24
`
`References 26
`
`BIO IndustryAnalysis I 5
`
`
`
`
`
`Introduction
`
`This study aimed to measure clinical development success rates to strengthen benchmarking metrics for drug development. To
`measure success rates for investigational drugs, we analyzed individual drug program phase transitions from January 1, 2006 to
`December 31, 2015. For the ten years studied, 9,985 transitions in the Biomedtracker database were analyzed. A phase transition
`is the movement out of a clinical phase — for example, advancing from Phase I to Phase II development, or being suspended
`after completion of Phase I development.
`
`These transitions occurred in 7,455 clinical drug development programs, across 1,103 companies (both large and small), making
`this the largest study of its kind. With this broad set of data, we aimed to capture the diversity in drug development across levels
`of novelty, molecular modalities, and disease indications.
`
`Only company-sponsored, FDA registration-enabling development programs were considered; investigator-sponsored studies
`were excluded from this analysis. A more detailed description ofthe data collection, composition, and analysis methodology
`are described at the end ofthis report under "Methods.”
`
`lndividual Phase transition success rates were determined by dividing the number that advanced to the next phase by the total
`number advanced and suspended. This “advanced and suspended” number is often referred to as “n” in this report, and should
`be taken into account when drawing conclusions from the success rate results.
`
`One of the key measures of success used in this report is the Likelihood of Approval (LOA) from Phase I. This LOA success rate
`is simply a multiplication of all four Phases success rates, a compounded probability calculation. For example, if each phase had
`a 50% chance of success, then the LOA from Phase I would be 0.5 x 0.5 x 0.5 x 0.5 = 6.25%.
`
`
`
`Lev:lei-)
`
`
`
`
`
`Phase Transition Success and Likelihood
`
`of Approval (LOA) - Overall
`
`Consistent with previous studies of drug development phase transition success rates, we found Phase II success rates to be far
`lower than any other phase.1 Phase I and Ill rates were substantially higher than Phase II, with Phase I slightly higher than Phase
`III. The highest success rate ofthe four development phases was the NDA/BLA filing phase.
`
`The Phase l transition success rate was 63.2% (n=3,582). As this Phase is typically conducted for safety testing and is not
`dependent on efficacy results for candidates to advance, it is common for this phase to have the highest success rate among the
`clinical phases across most categories analyzed in this report. Phase I success rates may also benefit from delayed reporting bias,
`as some larger companies may not deem failed Phase I programs as material and thereby not report them in the public domain.
`The Phase II transition success rate (30.7%, n=3,862) was substantially lower than Phase I, and the lowest ofthe four phases
`studied. As this is generally the first stage where proof-of—concept is deliberately tested in human subjects, Phase II consistently
`had the lowest success rate ofall phases. This is also the point in development where industry must decide whether to pursue
`the large, expensive Phase III studies and may decide to terminate development for multiple reasons including commercial
`viability. The second-lowest phase transition success rate was found in Phase III (58.1%, n=1,491). This is significant as most
`company-sponsored Phase III trials are the longest and most expensive trials to conduct.
`
`The probability of FDA approval after submitting a New Drug Application (NDA) or Biologic License Application (BLA), taking
`into account re-submissions, was 85.3% (n=1,050). Multiplying these individual phase components to obtain the compound
`probability of progressing from Phase Ito US. FDA approval (LOA) reveals that only 9.6% (n=9,985) of drug development
`programs successfully make it to market (Figure 1).
`
`90%
`
`80 0/0
`
`Probablity of Success
`
`85.3%
`
`ProbabilityofSuccess
`
`70 0/0
`
`hU'lalass53%
`
`3 0°/o
`
`1 00/0
`
`20°/o
`
`0 °/o
`
`Phase I to
`Phase II
`
`Phase II to Phase III to NDA/BLA to Phase I to
`Phase III
`NDA/BLA
`Approval
`Approval
`
`IA" Diseases, All Modalities
`
`Figure 1. Phase transition success rates and LOA from Phase I for all diseases, all modalities.
`
`BIO Industry Analysis I 7
`
`
`
`
`
`Phase Transition Success and Likelihood
`
`of Approval (LOA) - by Disease
`
`We segmented major disease areas according to the convention used by Biomedtracker, and categorized 21 major diseases
`and 558 indications for the 2006-2015 timeframe. For reporting at the disease area level, we analyzed only major diseases
`with more than 100 total transitions from Phase I to NDA/BLA approval. This resulted in 14 categorized disease areas: Allergy,
`Autoimmune, Cardiovascular, Endocrine, Hematology, Infectious disease, Gastroenterology (non-IBD), Metabolic, Neurology,
`Oncology, Ophthalmology, Psychiatry, Respiratory, and Urology. Disease areas with n <1OO were placed into the “Other” category.
`This includes Dermatology, Renal, Obstetrics, Rheumatology (for non-autoimmune indications), Dental, and Orthopedics.
`
`As can be seen in Figures 2a, there is a wide range of Likelihood of Approval (LOA) from Phase I. At the high end, Hematology
`towers over the other groups at 26.1% (n=283). A large portion of Hematology transitions came from Hemophilia, Anemia,
`and Blood Protein Deficiencies, Thrombocytopenia, and Hemostasis. Some ofthese Hemophilia indications had overall LOA
`that reached above 50%. This more than offset some ofthe weaker Hematology success rates that were observed in Venous
`Thromboembolism and Neutropenia. Hematology’s LOA from Phase I was 5x the success rate for Oncology, which at 5.1%
`(n=3,163) had the lowest of all the major disease areas.
`
`The next-highest LOA from Phase I under Hematology’s 26.1% was Infectious Disease with an impressive 19.1% (n=916).
`Five disease areas follow closely in the 14—17% range: Ophthalmology > Other > Metabolic > Gastroenterology > Allergy.
`Below 14% there is a third group of diseases that was slightly above the overall average of 9.6%: Endocrine > Respiratory >
`Urology > Autoimmune. Falling under the overall LOA of 9.6% was a fourth group made up of four disease areas: Neurology >
`Cardiovascular > Psychiatry > Oncology. The fact that Oncology and Neurology had the two highest n values while also having
`low LOA values suggests that these two disease categories are a significant factor in bringing down the overall industry LOA.
`
`Likelihood of Approval from Phase I
`
`26.1%
`
`
`
`
`19.1°lo
`
`o
`17.1 In 153%.
`15.3°/o 15.1°/0 14.7°/o
`
`13.2°/o 12.80/11
`
`11-4% 11.1%
`
`
`
`9.6%
`
`30%
`
`25°/o
`
`N‘3o\
`
`LOAfromPhaseI G°\°
`
`
`
`<‘5‘
`ee’
`v
`
`*0
`00‘
`(‘6
`q,
`
`06
`{5‘
`Q\
`06
`9v
`
`05‘
`o‘
`e‘
`
`$0
`6‘
`‘6‘
`‘9
`go
`
`-o“‘"
`5°
`80
`o
`‘0
`v
`
`05‘
`(0‘
`e9
`s‘
`
`.‘6
`6°
`so
`Q9
`
`09A
`o‘
`0"
`0
`
`*
`&°
`90
`40
`'s"
`‘6
`(5’
`
`0°“
`‘0‘
`e
`é“
`~b
`
`o"
`go
`‘90
`so
`
`a“?
`.90
`6‘
`
`0‘5‘
`o‘
`\6‘
`.59
`.5"
`0‘!
`
`’0‘"
`‘
`
`.
`$0
`5°
`a
`\"
`
`4r
`03‘
`\0
`0‘0
`s.
`Ge?
`($6
`(’9
`
`Figure 23. Chart of LOA from Phase I, displayed highest to lowest by disease area.
`
`8 I BlOlndustryAnalysis
`
`
`
`Phase Success
`
`Phase I to Phase II
`
`Phase II to Phase III
`
`Phase III to NDA/BLA
`
`Phase Success
`
`Phase Success
`
`NDA/BLA to Approval
`Advanced or
`Suspended
`
`Phase Success
`84.0%)
`88.7°/o
`77.5%
`88.4%
`77.8°/u
`92.3%
`93.8%
`86.0°/u
`94.6%
`85.7°/u
`86.0°/o
`
`83.2%
`84.2%
`87.9%
`82.4%
`
`Du
`43-A
`
`‘4
`
`O
`
`0“
`
`176
`
`NDA/BLA to Approval
`
`LOA n
`
`41-h
`
`l-'Luh—IN
`
`.9.
`
`Phase LOA
`84.0°/o
`88.7°/o
`77.5°/o
`88.4°/o
`77.8%
`92.3%
`93.8%
`86.0%
`94.6%
`85.7%
`86.0°/a
`
`83.2°/o
`84.2°/c
`87.9%
`82.4%:
`
`Suspended
`
`Suspended
`
`Hematologym— 33
`Infectious disease
`286
`
`56.6°/o
`42.7°/a
`
`44.6%
`Ophthalmology“— 101
`39.7%
`Other‘_ 116
`Metabolic“— 452%
`Gastroenterology"
`“ 35.7%
`llllergy
`325%
`242
`Endocrine
`40.1%
`196
`Respiratory
`29.1%
`Urology -I-_ 52
`327%
`Autoimmune
`319
`31.7%
`All Indications
`Neurology
`Cardiovascular
`psycho...
`Oncology
`
`29.7%
`24.1%
`23.7%
`24 6%
`
`465
`237
`169
`1416
`
`Likelihood of Approval
`
`Phase I to Approval
`
`Phase II to Approval
`
`Hematology
`Infectious disease
`Ophthalmology
`Other
`Metabolic
`Gastroenterology*
`Allergy
`Endocrine
`Respiratory
`Urology
`Autoimmune
`All Indications
`Neurology
`Cardiovascular
`Psychiatry
`Oncology
`
`Phase LOA
`35.7%
`27.5%
`20.1%
`24.4% ‘
`25.1%
`20.0%
`21.8%
`22.4%
`19.6%
`20.0%
`17.0%
`
`14.2°/o
`11.2°/o
`11.6°/a
`8.1%
`
`197
`
`201
`205
`146
`115
`I'D
`492
`278
`5?
`540
`
`842
`423
`297
`1941
`
`Advanced or
`Suspended
`
`LuVI
`
`H
`
`H
`
`.—-La
`
`H
`
`D
`
`11=- a:VI:9 ... H ...
`
`LOA n
`114
`283
`100
`
`Ln
`
`M
`
`D
`
`75.0%
`72.7%
`58.3%
`69.6%
`71.4°/o
`60.6%
`71.4%
`65.0%
`7 1 1%
`71.4°/o
`62.2%
`
`57.4%
`55.5%
`55.7%
`40.1%
`,.,o >'cc.,o<9’.
`
`Phase LOA
`63.0%
`64.5%
`45.2°/o
`61.5%
`55.6%
`55.9%
`67.0%
`55.9%
`67.3%
`61.2%
`53.5%
`
`47.80/0
`46.7°/o
`49.0%
`33.0°/o
`
`{ml
`
`y...IIN
`
`\l
`
`Figure 2b. Phase transition success and LOA by disease. Table of phase transition success and LOA by disease with corresponding n values.
`‘Advanced or Suspended’ refers to the total number of transitions used to calculate each success rate, with the n value noted in the text.
`The LOA n value is the total ‘Advanced or Suspended’ transitions of all phases used to calculate LOA. ‘Phase Success’ is the probability of
`successfully advancing to the next phase, whereas ‘Phase LOA’ is the probability of FDA approval for drugs from this phase of development.
`*Gastroenterology does not include IBD.
`
`BIO Industry Analysis I 9
`
`
`
`
`
`Phase I Transitiun Success Rates by Disease
`
`Success rates for Phase I ranged from 53.9% to 84.8%, with the average for all disease indications coming in at 63.2%. Looking
`at the distribution, we find that most disease area Phase I success rates cluster within +/—lO% of the overall Phase I success
`
`rate. Two disease areas were outliers, and they are both to the upside. Ophthalmology registered an 84.8% (n=66) success rate,
`which was substantially higher (by >20%) than the overall Phase l success rate. Gastroenterology programs also exhibited an
`above average rate of successfully overcoming initial clinical safety hurdles with a 75.6% (n=41) Phase I success rate.
`
`Phase II Transition Success Rates by Disease
`
`In every disease area, Phase II had the lowest transition success rate of the four phases. As shown in Figure 3, Phase II success
`rates ranged from a high of 56.6% (Hematology, n=83) to a low of23.7% (Psychiatry, n=169). Although it is widely known that
`drug program attrition is high in Phase ll, it is interesting to find that the rate of success can vary by 33% among disease groups.
`The only Phase II success rate above 50% was seen in Hematology, which largely explains how that indication attained the
`highest LOA from Phase l.
`
`Excluding Hematology, we can group the Phase II success rates into three clusters: success rates below the 30% overall
`
`success rate, those 31-36%, and those in the 40—45% range. Unlike what is observed in the LOA from Phase I, Oncology does
`not have the lowest success rate for Phase II. Cardiovascular and Psychiatry both registered slightly below the 25% success
`rate seen for Oncology.
`
`60°/o
`
`57%
`
`Probability of Phase II Success
`
`0 500/0
`*5
`0:
`3 400/0
`8
`3 30%
`m
`
`
`°
`
`45 /o
`450/0
`430/0
`40% 40%
`
`36% 330/
`°
`o 33 /o
`320/0 31°/o
`300/0
`29°/0
`
`
`25% 24% 24%
`
`
`
`20°/0
`
`100/0
`
`00/0
`
`A
`
`A
`
`.o
`
`°\<>9
`~o°\\
`°\°9
`O
`a \<“
`m‘
`to
`e
`s“
`o
`6‘
`o
`6°
`$0
`6
`0
`1‘00
`«5‘
`
`0
`
`0
`
`<
`
`dog as 0‘90 \05‘
`9
`a 05°
`so
`e
`n
`c
`00
`96
`60
`
`*
`
`A
`
`o\09
`0‘
`
`A
`
`o
`
`e
`
`A
`
`A
`
`<
`
`{56“
`o><b
`06,9 696 06,9
`{>00
`\éq {‘90
`\
`r
`’6“
`6“? 3 Q6
`00 f ‘6‘
`so
`6
`a”
`o
`'0
`9
`89
`Q.
`8
`Q
`0
`\\
`at
`v
`v.
`o
`
`HH
`
`0 8g
`
`a.
`
`Figure 3. Phase II transition success rates by disease area. Categories are listed from highest to lowest based on the probability of
`transitioning from Phase II to Phase III. *Gastroenterology does not include IBD.
`
`10 l BIO lndustryAnalysis
`
`
`
`Phase III Transition Success Rates by Disease
`
`For Phase lll transition success rates, Oncology was the outlier with the lowest transition success rate. As seen in Figure 4, the
`Phase III success rates for 14 specific disease areas clustered into two ranges: near 70% and 55-65%. This places Oncology
`into a group of its own atjust 40.1% (n=349).
`
`In addition to Oncology, Neurology, Psychiatry and Cardiovascular were also below the overall Phase lll success rate of 58.1%
`(n=1,491) at 57.4%, 55.7%, and 55.5%, respectively. Each of these areas included disease indications with large patient populations.
`Later in this report, we break down these high prevalence diseases and compare them with low prevalence disease areas.
`
`Probability of Phase III Success
`
`o
`73 /° 71% 71% 71% 71% 70%
`
`0/
`65 ° 62% 61%
`
`58% 58% 57% 56% 55%
`
`40°/o
`
`80% 75%
`
`70°/o
`
`was‘2?<5?
`
`PhaseIIISuccess Nwh<2ss;o\a5
`
`Figure 4. Phase III transition success rates by disease area. Categories are listed from highest to lowest based on the probability of transitioning
`from Phase II to NDAIBLA filing. *Gastroenterology does not include IBD.
`
`BIO Industry Analysis |
`
`11
`
`
`
`
`
`NDAIBLA Submission Success Rates
`
`NDA/BLA transition success rates (approval rates) for the disease areas listed in Figure 2b ranged from the low end of 77.5%
`(Ophthalmology) to a high of 94.6% (Respiratory). The distribution of rates (17.1%) were within the tightest range among the
`four phases analyzed in this report. These rates are the result of eventual success, not success on first review, meaning some
`programs may have as many as four Complete Response Letters (ORLs) and attempts at approval. This unrestricted time-frame
`and number of re-submissions pushes the overall success above 85% across all diseases.
`
`When looking at how many original NDA/BLA filings were approved on the first review by FDA, the rates are far from concentrated
`(Figure 5). In fact, Psychiatry had only a 37% chance offirst-Cycle approval vs. Oncology at nearly 80%. Although this is an
`extreme range, upon subsequent submissions and reviews, both of these disease areas ended up with 91% of original drug
`indication applications being approved. There was a large increase in cumulative success rates after the second submission,
`but only marginal increases after the third review.
`
`Time from filing to approval also varied by disease area. Neurology drugs took the longest to approve on average, at 2 years,
`while Oncology drugs were approved almost twice as fast at 1.1 years. Many Oncology drugs for unmet medical need may have
`benefited from expedited approval pathways and associated increased interactions with FDA such as Breakthrough Therapy
`and Accelerated Approval, contributing to the faster overall time to approval. As might be expected, calculating time to approval
`for all disease areas put the time to approval in the middle of these extremes, at 1.6 years.
`
`Disease Area
`
`% Approved on % Approved by % Ultimately
`lst Review
`2nd Review
`APPI‘OVed
`
`Filing to
`Approval Time
`(Years)
`
`Psychiatry
`
`Oncology
`
`Allergy
`
`Respiratory
`Cardiovascular
`
`Infectious disease
`
`Urology
`Autoimmune
`
`Metabolic
`
`Ophthalmology
`
`Hematology
`
`Gastroenterology
`Endocrine
`
`Neurology
`
`Figure 5. Time to FDA approval and percent approved by FDA for original NDAIBLA filings only. Data shown does not include
`supplemental applications.
`
`12 I BlOlndustryAnalysis
`
`
`
`
`
`Oncology and Non-Oncology Diseases
`
`Oncology drug development program transitions in the 2006-2015 period accounted for 31% of the 9,985 total transitions. With
`the lowest LOA from Phase I (5.1%, n=3,163), Oncology had an outsized effect on the overall industry success rate. To further
`understand this contribution, we compared phase transition success rates and LOA for non-oncology development programs
`against oncology development programs alone. Figure 6 shows Phase success rates and LOA from Phase l for oncology and
`non—oncology development programs. The LOA from Phase I across non-oncology indications was twice that for oncology alone,
`at 11.9% (n=6,822). Looking at individual phase transition success rates, it is clear that Phase III transition success rates were
`the reason Oncology ended up with the lowest overall success across our 14 disease categories. Oncology Phase lll success
`was 23% lower than Non-Oncology disease areas.
`
`1000/0
`
`900/0
`
`8 0 0/0
`
`7 0 0/0
`
`6 0 0/o
`
`5 0 0/0
`
`4 0°/o
`
`3 0°/o
`
`20°/o
`
`1 0°/o
`
`O 0/0
`
`ProbabilityofSuccess
`
`Probability of Success
`Oncology vs. Non-Oncology
`
`85.90/082.4°/o
`
`
`
`Phase I to
`Phase II
`
`Phase II to
`Phase III
`
`Phase III to NDA/BLA to
`NDA/BLA
`Approval
`
`Phase I to
`Approval
`
`Non-Oncology
`
`I Oncology
`
`Phase Success
`
`Phase I to Phase II
`
`Phase II to Phase III
`
`Phase III to NDA/BLA
`
`NDA/BLA to Approval
`
`Advanced or
`Suspended
`
`Oncology
`Non-Oncology
`
`1222
`2360
`
`Phase
`Success
`
`62.8%
`63.5%
`
`Advanced or
`Suspended
`
`1416
`2446
`
`Phase
`Success
`
`24.6%
`34.3%
`
`Advanced or
`Suspended
`
`349
`1142
`
`Phase
`Success
`
`40.1%
`63.7%
`
`Advanced or Phase Success
`Suspended
`
`82.4%
`85.9%
`
`Likelihood of Approval
`
`Phase I to Approval
`
`Phase II to Approval
`
`Phase III to Approval
`
`NDA/BLA to Approval
`
`2016
`
`LOA n
`
`Phase LOA
`
`oncology
`Non-Oncology
`
`3163
`6822
`
`5.1%
`11.9%
`
`LOA n
`
`1941
`4462
`
`Phase LOA
`
`LOA n
`
`Phase LOA
`
`Phase LOA
`
`8.1%
`18.7%
`
`525
`
`33.0%
`
`Figure 6. Oncology vs. Non-Oncology phase transition success rates and LOA. Top: Chart of LOA from Phase I. Bottom: Table of phase
`transition success rates and LOA for Oncology vs. Non-Oncology indications, with corresponding n values. ‘Advanced or Suspended’
`refers to the total number of transitions used to calculate each success rate, with the :1 value noted in the text. The LOA n value is
`the total ‘Advanced or Suspended’ transitions of all phases used to calculate LOA. ‘Phase Success’ is the probability of successfully
`advancing to the next phase, whereas ‘Phase LOA’ is the probability of FDA approval for drugs in this phase of development.
`
`BlOlndUstryAnalysis | 13
`
`
`
`
`
`Oncology Sub-Indication Phase Transition Success Rates and LOA
`
`Oncology drugs were further categorized into two main types of cancer: solid tumors and hematological cancers. Solid tumors
`had twice as many transitions in the data set (2.283 vs. 805), but only halfthe LOA from Phase I vs. hematological cancers (4.0%
`vs. 8.1%). These are shown in Figure 7 in more detail.
`
`0 0/0
`
`1 0 0/0
`
`200/0
`
`3 0 0/0
`
`4 0 0/0
`
`5 0 0/0
`
`600/0
`
`7 0 0/0
`
`Phase III Success Rate
`
`Oncology Indications
`
`Hematologic Cancers
`CLL/SLL - NHL
`
`Multiple Myeloma
`Indolent NHL
`AML
`
`Solid Tumors
`
`Renal Cell Cancer
`
`Breast Cancer
`Melanoma
`Prostate Cancer
`
`Colorectal Cancer
`
`NSCLC
`
`Ovarian Cancer
`
`Gastric Cancer
`Pancreatic Cancer
`
`86.4%
`
`28.7%
`
`52.6%
`
`8
`
`59
`
`79-63%
`86.4%
`
`Phase Success
`
`Phase I to Phase II
`Advanced or
`Suspended
`1222
`
`Oncology
`Solid
`Hematologic— 61.8%
`
`62.8%
`
`Phase II to Phase III
`Advanced or
`Suspended
`1416
`
`Phase III to NDA/BLA
`Advanced or
`Suspended
`
`NDA/ BLA to Approval
`Advanced or
`Suspended
`
`Likelihood of Approval
`
`Phase I to Approval
`
`Phase II to Approval
`
`Phase III to Approval
`
`NDA/BLA to Approval
`
`Oncology
`Solid
`Hematologic
`
`3163
`2283
`
`Phase LOA
`
`1941
`
`Phase LOA
`8.1%
`6.3%
`13.1%
`
`LOA n
`525
`36
`
`Phase LOA
`33.0%
`27.3%
`45.4%
`
`LOA n
`176
`108
`
`Phase LOA
`824%
`79.6%
`
`Figure 7. Phase transition success rates and LOA for Oncology indications with corresponding n values. ‘Advanced or Suspended'
`refers to the total number of transitions used to calculate each success rate. with the n value noted in the text. The LOA n value is
`the total ‘Advanced or Suspended' transitions of all phases used to calculate LOA. ‘Phase Success’ is the probability of successfully
`advancing to the next phase. whereas ‘Phaee LOA’ is the probability of FDA approval for drugs in this phase of development.
`
`14 l BIO Industry Analysis
`
`
`
`
`
`Phase ll transition success rates by sub-indication tended to range close to the overall 25% Oncology calculation for Phase ll, +/-10%.
`
`Narrowing in on Phase ”I transition success rates, only 34.2% ofthe 260 drug programs in solid tumor cancers were deemed
`sufficiently successful to file an NDA/BLA with the FDA. This was the underlying cause for the 2x difference we see in overall LOA,
`as the hematological cancer programs recorded a 52.6% success rate in Phase I”. Since Phase III was identified as the weakest
`phase for Oncology, Phase III transition success rates for a number of major oncology sub—indications were included in Figure 7.
`
`The solid tumor Phase “I transition success rate (34.2%, n=260) ended up as the lowest for any of the major disease categories
`studied. Solid tumor drugs for pancreatic cancer seemed to have the toughest challenges in Phase III studies (13.3%, n=15).
`However, Phase III success rates for Ovarian and Gastric cancers also fell below 30%.
`
`Hematological cancer Phase III transition success rates benefited from transition successes in CLL/SLL (66.7%, n=12) and
`MM (63.7%, n=11). ALL, Hodgkin’s and CML had Phase III success rates of 100% but only had fewer than five completed clinical
`development programs each (data not shown). Only AML (36.4%, n=11) came in below 40% for Phase III, which helped the overall
`hematologic cancer Phase III success rate remain above 50%.
`
`The NDA/BLA to approval success rate for all hematological cancers (86.4%, n=59) was impacted positively by Multiple Myeloma,
`ALL, and CML success, as each had more than five completed filings and 100% approval rates. The NDA/BLA success rate for
`all sold tumors was lower at 79.6% (n=108).
`
`Abbreviated cancer indications:
`
`ALL
`
`- Acute Lymphocytic Leukemia
`
`AML
`
`- Acute Myelogenous Leukemia
`
`CLL
`
`- Chronic Lymphocytic Leukemia
`
`CML
`
`- Chronic myelogenous Leukemia
`
`MM
`
`— Multiple Myeloma
`
`NHL
`
`- Non-Hodgkin’s Lymphoma
`
`NSCLC - Non-Small Cell Lung Cancer
`
`SLL
`
`- Small Lymphocytic Lymphoma
`
`BIO Industry Analysis
`
`'r"
`
`
`
`
`
`Rare Diseases and Chronic High Prevalence Diseases
`
`in recent years, there has been an increase in funding for companies focused on rare diseases.2 This is welcome news as there
`are reportedly 7,000 rare diseases and most do not have an approved therapeutic treatment.4 One question that is often asked
`is ifthe probabilities of success are any better for rare diseases, especially for those in which a particular defective gene has
`been confirmed as the sole contributor. On the other extreme, we have observed less venture funding for high prevalence,
`chronic diseases.2 The question we wanted to explore is whether investors may have scaled back funding because there is a
`higher hurdle to developing and gaining approval for medicines that treat highly prevalent conditions.
`
`Probability of Success
`Rare Disease and High Prevalence Diseases
`
`890/0
`
`870/0
`
`850/0
`
`1000/0
`900/0
`
`8 0 0/0
`
`ProbabilityofSuccess
`
`7 O 0/0
`
`6 O 0/0
`
`5 0 0/o
`
`4 O 0/0
`
`3 0 0/0
`
`2 O 0/0
`
`1 0 0/0
`
`0 0/0
`
`Phase I to
`Phase II
`
`Phase II to
`Phase III
`
`Phase III to
`NDA/BLA
`
`NDA/BLA to
`Approval
`
`Phase I to
`Approval
`
`IAII Diseases
`
`IChronic High Prevalence
`
`lRare Diseases
`
`Phase Success
`
`Phase I to Phase II
`
`Phase II to Phase III
`
`Phase III to NDA/BLA
`
`NDA/BLA to Approval
`
`A..D.seases
`
`Rare Diseases
`
`Advanced or Phase Success
`Suspended
`3582
`732
`150
`
`76.0%
`
`Advanced or
`Suspended
`3862
`26
`168
`
`Phase Success
`
`Advanced or
`Suspended
`
`Phase
`Success
`
`Advanced or
`Suspended
`
`Phase Success
`
`50.6%
`
`10
`
`1
`
`73.6%
`
`El
`
`892%
`
`Likelihood of Approval
`
`Phase I to Approval
`
`Phase II to Approval
`
`Phase III to Approval
`
`NDA/BLA to Approval
`
`89.2%
`
`All Diseases
`Chronic High Prevalence
`Rare Diseases
`
`LOA n
`9985
`1922
`521
`
`Phase LOA
`9.6%
`8.7%
`25.3%
`
`LOA n
`6403
`1190
`371
`
`Phase LOA
`15.3%
`14.9%
`33.3%
`
`LOA n
`2541
`464
`203
`
`Phase LOA
`49.6%
`53.7%
`65.7%
`
`LOA n
`1050
`196
`
`Phase LOA
`85.3%
`87.2%
`
`Figure 8. Non-Oncology Rare disease and high prevalence, chronic disease. Top: Chart of phase transition success rates and LOA from
`Phase I, for rare and chronic, high prevalence. Bottom: Table of phase transition success and LOA by disease with corresponding n values.
`‘Advanced or Suepended’ refers to the total number of transitions used to calculate each success rate, with the n value noted in the text.
`The LOA n value is the total ‘Advanced or Suspended’ transitions of all phases used to calculate LOA. ‘Phase Success’ is the probability of
`suceesafially advancing to the next phase, whereas ‘Phase LOA’ is the probability of FDA approval for drugs in this phase of development.
`
`16 l BlOlndustryAnalysis
`
`
`
`
`
`We isolated rare disease programs in the Biomedtracker database by first identifying all prior FDA Orphan-designated indications,
`then pooling all drug programs in these indications, regardless of whether they obtained FDA’s Orphan designation. A|| Oncology
`indications were removed to make this rare disease analysis more concentrated on inborn genetic disorders. For chronic
`diseases, we first obtained a list of conditions from the Center for Medicare & Medicaid Services (CMS) Chronic Conditions
`
`Data Warehouse (COW). We removed any cancer indications, then identified those disease with > 1 million patients afflicted
`with the disease in the United States.
`
`With programs from both groups identified, we compared phase transition success rates and LOA as shown in Figure 8. At 25.3%,
`the overall LOA from Phase I for Non-Oncology rare diseases was 2.6x higher than the LOA for all diseases and 3x higher than the
`8.7% LOA for chronic, high prevalence diseases. Chronic, high prevalence diseases accounted for almost 20% of the transitions,
`and Oncology, 31% of the transitions. The combined weighting of 50% of the dataset for these two categories with low LOAs
`(Chronic, high prevalence disease, 8.7% and Oncology, 5.7%) contributed significantly to the low overall industry LOA of 9.6%.
`
`All four transition success rates were higher for the rare disease group than the overall dataset. The largest difference between
`was found in Phase II transition success rates (50.6% for rare disease vs 30.7% overall). Phase III and Phase I each had higher
`transition success rates by at least 10%.
`
`Chronic, high prevalence diseases transition success rates were lower in Phase I (58.7% vs 63.2%) and Phase II (27.7% vs. 30.7%)
`vs. the overall dataset. The opposite was seen in Phase III and NDA/BLA, where slightly higher rates were observed: 61.6% vs.
`58.1% for Phase Ill and 87.2% vs. 85.3% for NDA/BLA.
`
`As the chronic, high prevalence group in this study does not include Oncology, we compared these results with the Non-Oncology
`rates found in Figure 6. Although the chronic, high prevalence Phase III transition success rate is higher than that seen with all
`diseases, it is lower than the Non—Oncology transition success rates (61.6% vs. 63.7%). Furthermore, the LOA of 11.9% (n=6,822)
`for all Non-Oncology indications is 3% higher than for chronic, high prevalence diseases (8.7%), suggesting that these chronic
`indications are negatively impacting overall success rates outside Oncology.
`
`BIO lndustryAnalysis ‘
`
`!"
`
`
`
`
`
`Patient Selection Biomarker Programs
`
`The use of biomarkers as inclusion or exclusion criteria, or ‘selection biomarkers’, for enrolling patients into clinical studies
`has increased dramatically since the sequencing ofthe human genome. We identified 512 phase transitions out of 9,985
`(5%) that incorporated a selection biomarker for patient stratification. This was accomplished by mapping the NOT numbers
`(clinicaltrialgov identifier) from Amplion’s BiomarkerBase to programs in the Biomedtracker transition database.e For
`programs that filed an NDA/BLA, we only included filings that used selection biomarkers in the Phase III study design.
`
`The LOA from Phase i can be found in Figure 9. The benefit from selection biomarker use raises the LOA from Phase I to one
`in four compared to less than one in 10 when no selection biomarker was used.
`
`Probability of Success
`With