EUTECNIQ Sciences

See every page on one mapArrow keys move through results · Enter opens · Esc closes

Academy · Course 2 Primary sources Explorers

At least 28,187 plant species are recorded as medicinal, and only a small share carry the evidence a U.S. product needs

Fewer than 16% of medicinal plant species are cited in a regulatory publication. This course follows that gap through U.S. law and FDA data: identity, safety evidence, the new dietary ingredient rule, the drug route, patents and access.

Level
Introductory to intermediate
Reading time
About 25 minutes
Builds on
History of botanical medicine
Tools
2 explorers, 1 Sankey, 1 heat map

Status. EUTECNIQ is a company in formation with no laboratory, data, customers or patents. This course teaches published facts, not legal advice.

28,187plant species recorded as medicinal (Kew, 2017)
15.9%of those species cited in a medicinal regulatory publication
1,372notifications on FDA’s public new-dietary-ingredient list since 1995
4botanical drugs approved by FDA, 2006 to 2023
At a glance

Of 342,953 accepted vascular plant species, 28,187 have a recorded medicinal use and 4,478 appear in a medicinal regulatory publication

How the species count narrows from all vascular plants to those cited by regulatorsKew data
World Checklist of Vascular Plants (342,953 accepted species, April 16, 2021); Kew State of the World's Plants 2017 and Medicinal Plant Names Services (28,187 medicinal species; 4,478 cited in a medicinal regulatory publication). The two counts come from different Kew sources and dates.
Data table: the three counts
Species counts in the flow
StepSpeciesShare of the step before
Accepted vascular plant species342,953–
Recorded as of medicinal use28,1878.2%
Cited in a medicinal regulatory publication4,47815.9%

You will learn

  1. Read the medicinal-plant funnel from published counts and say what each step measures.
  2. Explain why names, plant parts and chemistry make identity the first barrier.
  3. Describe the new dietary ingredient (NDI) rule and what FDA's public notification list shows.
  4. Compare the effort behind supplement, food-ingredient and drug evidence using sourced estimates.
  5. Explain how the drug-first clause, patent limits, CITES and access-and-benefit-sharing narrow the field.
  6. Walk one candidate through seven gates and find the first question that stops it.
The funnel

Fewer than one in six recorded medicinal species appears in any medicinal regulatory publication

From accepted plant species to species named in a regulatory text, each step loses most of the one beforeKew data
Funnel on a logarithmic scale: 342,953 accepted vascular plant species, at least 28,187 recorded as medicinal (8.2 percent), and 4,478 cited in a medicinal regulatory publication (15.9 percent of medicinal species); about 1,280 medicinal species are under CITES protectionAccepted vascular plant speciesWorld Checklist of Vascular Plants, April 2021Accepted vascular plant species: 342,953342,953Recorded as being of medicinal useKew Medicinal Plant Names Services, 2017Recorded as being of medicinal use: at least 28,187at least 28,187 · 8.2% of all speciesCited in a medicinal regulatory publicationKew, 2017: pharmacopoeias and other regulatorytextsCited in a medicinal regulatory publication: 4,4784,478 · 15.9% of medicinal speciesMedicinal species under CITES protectionKew, 2017; a subset of the medicinal speciesAbout 1,280 of the 28,187 medicinal species are under CITES protectionabout 1,280 · 4.5% of medicinal species1001,00010,000100,0001,000,000Number of species (logarithmic scale)Funnel: 342,953 accepted vascular plant species; at least 28,187 recorded as medicinal; 4,478 cited in a medicinal regulatory publication; about 1,280 medicinal species under CITES protectionAccepted vascular plant speciesWorld Checklist of Vascular Plants, April 2021Accepted vascular plant species: 342,953342,953Recorded as being of medicinal useKew Medicinal Plant Names Services, 2017Recorded as being of medicinal use: at least 28,187at least 28,1878.2% of accepted speciesCited in a medicinal regulatory publicationKew, 2017: pharmacopoeias and other regulatory textsCited in a medicinal regulatory publication: 4,4784,47815.9% of medicinal speciesMedicinal species under CITES protectionKew, 2017; subset of the medicinal speciesMedicinal species under CITES protection: about 1,280about 1,2804.5% of medicinal species1001,00010,000100,0001MSpecies, logarithmic scale
Govaerts R et al. Sci Data 2021;8:215 (342,953 accepted species, April 16, 2021). Allkin B. Useful plants: medicines. In: Willis KJ, ed. State of the World's Plants 2017. Royal Botanic Gardens, Kew; 2017 (28,187; 4,478; c. 1,280 under CITES).
Data and limits of the funnel
StepSpeciesShare of previous stepSource
Accepted vascular plant species342,953–World Checklist of Vascular Plants, April 16, 2021
Recorded as being of medicinal use28,1878.2%Kew Medicinal Plant Names Services, 2017
Cited in a medicinal regulatory publication4,47815.9%Kew, 2017
Medicinal species under CITES protectionc. 1,2804.5% of medicinal speciesKew, 2017

Kew calls 28,187 “probably a very conservative figure.” A regulatory publication includes pharmacopoeias from many countries; being cited is not U.S. authorization. Species are not ingredients: one species can yield several parts and extracts. We found no verified count of U.S. pharmacopeial botanical monographs and do not show one.

Each year FDA receives thousands of label notifications, dozens of new-ingredient notifications and about twenty botanical drug applications to studyFDA data
Per year on a logarithmic scale: about 3,690 structure/function notifications, about 55 new dietary ingredient notifications, about 20 botanical INDs, and 4 botanical drug approvals in 18 yearsStructure/function notificationsFDA estimate, July 2025FDA estimate of annual 21 CFR 101.93 notifications (90 FR 35694)≈3,690 per yearNew dietary ingredient notificationsFDA estimate, January 2024FDA estimate of annual NDINs under 21 CFR 190.6 (89 FR 2957)≈55 per yearBotanical INDs received by CDER739 from March 1984 to December 2020739 botanical INDs over about 36.8 years (Park et al. 2025)≈20 per yearBotanical drugs approved4 approvals, 2006 to 2023Veregen 2006, crofelemer 2012, NexoBrid 2022, Filsuvez 20234 in 18 years0.11101001,00010,000Submissions or approvals per year (logarithmic scale)
FDA burden estimates: 90 FR 35694 (July 29, 2025) and 89 FR 2957 (January 17, 2024). Park JK et al. Ther Innov Regul Sci 2025;59:1129–1137 (739 botanical INDs, March 1984 to December 2020). Approvals: Drugs@FDA. Rates are per year; the units differ by row.
Data for the yearly rates
ItemCountPeriodPer year
Structure/function notifications, 21 CFR 101.933,690FDA annual estimate3,690
New dietary ingredient notifications, 21 CFR 190.655FDA annual estimate55
Botanical INDs received by CDER739March 1984 to December 2020 (about 36.8 years)≈20
Botanical drugs approved42006 to 20230.22
Explanation, details and sources

Each step counts something different, so read the ratio between steps, not the drop alone. The share at step k is fk = nk ÷ nk−1: 28,187 ÷ 342,953 = 8.2%, and 4,478 ÷ 28,187 = 15.9%.

Barrier 1 · Identity

Identity comes first, because one trade name can cover 15 species and one species can carry 15 names

Names fan out both ways: “ginseng” covers 15 species, and a medicinal species has 15 alternative names on averageKew data
Two fans: the trade name ginseng refers to 15 different plant species; on average each medicinal species has 15 alternative names; 143 databases and publications cite 415,180 unique names for plant-based medicinesOne trade name, many species“ginseng”Species 1 of 15 sold under the trade name ginsengSpecies 2 of 15 sold under the trade name ginsengSpecies 3 of 15 sold under the trade name ginsengSpecies 4 of 15 sold under the trade name ginsengSpecies 5 of 15 sold under the trade name ginsengSpecies 6 of 15 sold under the trade name ginsengSpecies 7 of 15 sold under the trade name ginsengSpecies 8 of 15 sold under the trade name ginsengSpecies 9 of 15 sold under the trade name ginsengSpecies 10 of 15 sold under the trade name ginsengSpecies 11 of 15 sold under the trade name ginsengSpecies 12 of 15 sold under the trade name ginsengSpecies 13 of 15 sold under the trade name ginsengSpecies 14 of 15 sold under the trade name ginsengSpecies 15 of 15 sold under the trade name ginseng15 speciesshare the nameOne species, many names1 speciesAlternative name 1: the average medicinal species has 15Alternative name 2: the average medicinal species has 15Alternative name 3: the average medicinal species has 15Alternative name 4: the average medicinal species has 15Alternative name 5: the average medicinal species has 15Alternative name 6: the average medicinal species has 15Alternative name 7: the average medicinal species has 15Alternative name 8: the average medicinal species has 15Alternative name 9: the average medicinal species has 15Alternative name 10: the average medicinal species has 15Alternative name 11: the average medicinal species has 15Alternative name 12: the average medicinal species has 15Alternative name 13: the average medicinal species has 15Alternative name 14: the average medicinal species has 15Alternative name 15: the average medicinal species has 1515 nameson average143 databases and publications cite 415,180 unique names for plant-based medicines (Kew, 2017)
Allkin B, State of the World's Plants 2017, Kew (415,180 names in 143 databases and publications). Dots are counts, not named species. U.S. law narrows this name: a food is misbranded if represented as ginseng “unless it is an herb or herbal ingredient derived from a plant classified within the genus Panax” (21 U.S.C. 343(u)).
FDA investigators cited missing identity tests 66 times in fiscal 2025, up from 15 in fiscal 2023FDA data
Grouped bars of FDA Form 483 citations by fiscal year. Identity test, 21 CFR 111.75(a)(1)(i): 22, 15, 39, 66 for FY2022 to FY2025. Supplier qualification, 111.75(a)(2)(ii)(A): 25, 18, 29, 450204060Citations per fiscal yearFY2022, Identity test, 111.75(a)(1)(i): 2222FY2022, Supplier qualification, 111.75(a)(2)(ii)(A): 2525FY2022FY2023, Identity test, 111.75(a)(1)(i): 1515FY2023, Supplier qualification, 111.75(a)(2)(ii)(A): 1818FY2023FY2024, Identity test, 111.75(a)(1)(i): 3939FY2024, Supplier qualification, 111.75(a)(2)(ii)(A): 2929FY2024FY2025, Identity test, 111.75(a)(1)(i): 6666FY2025, Supplier qualification, 111.75(a)(2)(ii)(A): 4545FY2025Identity test of each dietaryingredient, 111.75(a)(1)(i)Supplier certificate relied on withoutqualification, 111.75(a)(2)(ii)(A)Identity-test citations rose from 15 inFY2023 to 66 in FY2025Counts, not rates: the number of inspectionsdiffers from year to year
U.S. FDA, Inspection observations, FY2022–FY2025 (Form 483 citations under 21 CFR Part 111), as tabulated on Competitive landscape.
Data: Form 483 citations by fiscal year
Fiscal yearIdentity test, 111.75(a)(1)(i)Supplier qualification, 111.75(a)(2)(ii)(A)All Part 111
FY202222251,218
FY202315181,173
FY202439291,583
FY202566451,938
Explanation, details and sources

A notification must name a botanical by “the Latin binomial name, including the author” (21 CFR 190.6(b)(2)), and a supplement maker must run at least one identity test on each dietary ingredient (21 CFR 111.75(a)(1)(i)). Course 3 teaches the methods.

Barrier 2 · New dietary ingredients

With no official list of pre-1994 ingredients, each firm documents its own status, and FDA's public list holds 1,372 notifications

Publicly listed NDINs ran between 2 and 93 a year, near FDA's planning estimate of 55FDA data
Bars of publicly listed new dietary ingredient notifications by year of submission, 1995 to 2025; the yearly count ranges from 2 in 1995 to 93 in 2017, against FDA's planning estimate of 55 per year020406080100Notifications submitted per year1995: 2 notifications listed219951996: 8 notifications listed1997: 14 notifications listed1998: 19 notifications listed1999: 24 notifications listed2000: 31 notifications listed20002001: 20 notifications listed2002: 53 notifications listed2003: 64 notifications listed2004: 43 notifications listed2005: 57 notifications listed20052006: 64 notifications listed2007: 73 notifications listed2008: 40 notifications listed2009: 47 notifications listed2010: 53 notifications listed20102011: 49 notifications listed2012: 42 notifications listed2013: 33 notifications listed2014: 40 notifications listed2015: 48 notifications listed20152016: 60 notifications listed2017: 93 notifications listed932018: 42 notifications listed2019: 33 notifications listed2020: 54 notifications listed20202021: 47 notifications listed2022: 33 notifications listed2023: 45 notifications listed2024: 59 notifications listed2025: 71 notifications listed712025FDA planning estimate: 55 per year
U.S. FDA, “List of NDINs 1995 to the Present” (spreadsheet downloaded October 23, 2026), counted by year of submission; 89 FR 2957 for the estimate. All ingredient types, not only botanicals.
Data: NDINs listed by year of submission
YearsNotifications per year, in order
1995–20042, 8, 14, 19, 24, 31, 20, 53, 64, 43
2005–201457, 64, 73, 40, 47, 53, 49, 42, 33, 40
2015–202548, 60, 93, 42, 33, 54, 47, 33, 45, 59, 71

The list has 1,372 rows numbered up to 1,460 (first submission July 7, 1995; last listed February 11, 2026). Two rows have a typing error in the submission date and are left out of the yearly counts; nine rows dated 2026 are not shown. FDA lists “publicly displayable” notifications it has reviewed, without the type of response.

Explore the NDIN record FDA data

Range summary, years y₀ to y₁N = Σ ny · mean = N ÷ years
Notifications listed537
Mean per year53.7
Median days to FDA's response68
Responses after day 756.9%

2016 to 2025: 537 notifications, a mean of 53.7 a year, against FDA's estimate of 3,690 structure/function notices a year (about 69 to 1).

Notifications by year; the selected range is darkerFDA data
Response time is the date of FDA's response minus the submission date in FDA's list.
Method, assumptions and limits
What it calculates
For the chosen calendar years: the number of notifications in FDA's public list, the mean per year, the median number of days from the listed submission date to FDA's response, and the share of responses dated more than 75 days after submission.
Equations
N = Σ ny; mean = N ÷ number of years; the median is the middle value of the sorted response times (the mean of the two middle values for an even count).
Variables and units
y0, y1: calendar years, 1995–2025 (if reversed they are swapped). Response time: days. Two rows whose response predates submission are excluded from response times.
What it does not establish
The list holds only publicly displayable notifications and does not record whether FDA objected. A response after day 75 can follow an amendment that set a new filing date (21 CFR 190.6(d)); it is not a measure of FDA delay.
Verification
Six ranges are checked against an independent Python count of the same spreadsheet (tests/numerics/ref/academy_untapped-botanicals.py).
References
U.S. FDA, Submitted 75-Day Premarket Notifications for New Dietary Ingredients (list downloaded October 23, 2026); 21 CFR 190.6; 89 FR 2957; 90 FR 35694.
Explanation, details and sources

A new dietary ingredient is one “not marketed in the United States before October 15, 1994” (FD&C Act § 413(d)). FDA states: “There is no authoritative list of dietary ingredients that were marketed in dietary supplements before October 15, 1994.” It held a public meeting on such a list on October 3, 2017. Details are on NDIN strategy.

The safety basis is the costly part. When a notification relies on history of use alone, FDA “generally recommends 25 years of widespread use to be the minimum” (final guidance, October 2026); otherwise toxicology studies fill the gap between historical and proposed intake.

The explorer opens on 2016–2025, when FDA's median response came on day 68. Set the range to 1995–2025 and the median is 71 days, the all-years figure on NDIN strategy.

Barrier 3 · Cost of evidence

Sourced estimates put a supplement notice in hours of paperwork and the evidence for a new drug in billions of dollars

Preparation effort rises from under an hour to 180 hours; drug evidence costs millions per trial and billions per approvalFDA estimates Peer-reviewed
Two log-scale ladders. FDA's paperwork estimates per submission: 0.75 hours for a structure/function notification, 20 hours for an NDIN, 180 hours for a GRAS notice. Published drug-evidence costs: median pivotal trial 19.0 million dollars; 1.40 billion out of pocket and 2.56 billion capitalized per approved new drugFDA estimates of the time to prepare a submission from data already heldStructure/function notification21 CFR 101.93; FDA, 2025Structure/function notification: 0.75 hours0.75 hoursNew dietary ingredient notification21 CFR 190.6; FDA, 2024New dietary ingredient notification: 20 hours20 hoursGRAS notice for a food substanceFDA proposed rule, 2026GRAS notice for a food substance: 180 hours180 hours0.11101001,000Hours per submission (logarithmic scale)Published estimates of the cost of drug evidencePivotal efficacy trial, median138 trials behind 59 drugs approved 2015–2016Pivotal efficacy trial, median: $19.0 million$19.0 millionOut-of-pocket R&D per approved new drug106 drugs; 2013 dollarsOut-of-pocket R&D per approved new drug: $1.40 billion$1.40 billionCapitalized R&D per approved new drugsame sample, 10.5% real cost of capitalCapitalized R&D per approved new drug: $2.56 billion$2.56 billion$1M$10M$100M$1B$10BUS dollars per item (logarithmic scale)Paperwork estimates: 0.75 hours, 20 hours and 180 hours per submission. Drug evidence costs: 19.0 million dollars median pivotal trial; 1.40 billion and 2.56 billion dollars per approved new drugFDA preparation time, hoursStructure/function notification21 CFR 101.93; FDA, 2025Structure/function notification: 0.75 hours0.75 hoursNew dietary ingredient notification21 CFR 190.6; FDA, 2024New dietary ingredient notification: 20 hours20 hoursGRAS notice for a food substanceFDA proposed rule, 2026GRAS notice for a food substance: 180 hours180 hours0.11101001,000Drug evidence cost, US dollarsPivotal efficacy trial, median138 trials behind 59 drugs approved 2015–2016Pivotal efficacy trial, median: $19.0 million$19.0 millionOut-of-pocket R&D per approved new drug106 drugs; 2013 dollarsOut-of-pocket R&D per approved new drug: $1.40 billion$1.40 billionCapitalized R&D per approved new drugsame sample, 10.5% real cost of capitalCapitalized R&D per approved new drug: $2.56 billion$2.56 billion1M10M100M1B10B
90 FR 35694 (2025); 89 FR 2957 (2024); 91 FR 51834 (2026, proposed rule). Moore TJ et al. JAMA Intern Med 2018;178:1451–1457. DiMasi JA, Grabowski HG, Hansen RW. J Health Econ 2016;47:20–33 (2013 dollars). Dollar figures are for drugs in general, not botanicals.
Data for the ladder
ItemEstimateBasis
Structure/function notification0.75 hoursFDA, 3,690 notices a year
NDI notification20 hoursFDA, extracting and summarizing information already in the company's files
GRAS notice180 hoursFDA preliminary regulatory impact analysis, August 2026
Pivotal efficacy trial$19.0 million (median; IQR 12.2–33.1)138 trials behind 59 drugs approved 2015–2016
Out-of-pocket R&D per approved new drug$1,395 million106 drugs from 10 firms; 2013 dollars
Capitalized R&D per approved new drug$2,558 millionSame sample, 10.5% real cost of capital
Explanation, details and sources

FDA's hour estimates cover preparing a submission from data the firm already holds; they exclude generating studies. Drug estimates capitalize spending to approval and absorb failures: C = Σ ct(1 + r)T−t, with ct the spending in year t, T the approval year and r = 10.5% a year in DiMasi et al. Worked example: $1 spent ten years before approval counts as 1.10510 = $2.71 at approval.

Barrier 4 · The drug route

FDA received 739 botanical INDs from 1984 to 2020 and has approved four botanical drugs, each with a composition specification

Most botanical INDs were allowed to proceed at 30 days, and about half were later withdrawn or terminatedFDA authors, peer-reviewed
Heat map of the share of botanical INDs in each initial FDA action and each status in December 2020, for oncologic (254), non-oncologic (485) and all (739) INDsInitial 30-day actionOncologic (n = 254)Non-oncologic (n = 485)All (n = 739)Safe to proceedSafe to proceed, oncologic: 175 of 254 (69%)Safe to proceed, non-oncologic: 282 of 485 (58%)Safe to proceed, all: 457 of 739 (62%)Clinical holdClinical hold, oncologic: 56 of 254 (22%)Clinical hold, non-oncologic: 111 of 485 (23%)Clinical hold, all: 167 of 739 (23%)WithdrawalWithdrawal, oncologic: 12 of 254 (5%)Withdrawal, non-oncologic: 14 of 485 (3%)Withdrawal, all: 26 of 739 (4%)Other initial actionsOther initial actions, oncologic: 11 of 254 (4%)Other initial actions, non-oncologic: 78 of 485 (16%)Other initial actions, all: 89 of 739 (12%)175 (69%)282 (58%)457 (62%)56 (22%)111 (23%)167 (23%)12 (5%)14 (3%)26 (4%)11 (4%)78 (16%)89 (12%)Status in December 2020ActiveActive, oncologic: 82 of 254 (32%)Active, non-oncologic: 171 of 485 (35%)Active, all: 253 of 739 (34%)On holdOn hold, oncologic: 7 of 254 (3%)On hold, non-oncologic: 33 of 485 (7%)On hold, all: 40 of 739 (5%)Withdrawn or terminatedWithdrawn or terminated, oncologic: 137 of 254 (54%)Withdrawn or terminated, non-oncologic: 237 of 485 (49%)Withdrawn or terminated, all: 374 of 739 (51%)InactiveInactive, oncologic: 24 of 254 (9%)Inactive, non-oncologic: 31 of 485 (6%)Inactive, all: 55 of 739 (7%)OtherOther, oncologic: 4 of 254 (2%)Other, non-oncologic: 13 of 485 (3%)Other, all: 17 of 739 (2%)82 (32%)171 (35%)253 (34%)7 (3%)33 (7%)40 (5%)137 (54%)237 (49%)374 (51%)24 (9%)31 (6%)55 (7%)4 (2%)13 (3%)17 (2%)Share of INDs, %070Counts in each cell;shares of the column total
Park JK, Lee D, Rui L, Gao X, Furness MS, Wu C. Ther Innov Regul Sci 2025;59(5):1129–1137, Table 1. Shares for “All” computed here from the counts.
Data: botanical INDs by action and status
CategoryOncologic (254)Non-oncologic (485)All (739)
Safe to proceed at 30 days175282457
Clinical hold at 30 days56111167
Withdrawal at 30 days121426
Other initial actions117889
Active, December 202082171253
On hold, December 202073340
Withdrawn or terminated137237374
Inactive243155
Other status41317
Four plants, four drug substances, four approved usesFDA labels
Four rows from plant to approved drug: green tea leaf to Veregen (2006), Croton lechleri latex to crofelemer (2012), pineapple stem to NexoBrid (2022), birch bark to Filsuvez (2023), each with its drug substance and approved usePlant and partDrug substanceProduct and FDA actionApproved useGreen tea leafCamellia sinensisSinecatechins, a partially purifiedwater-extract fractionVeregen ointment 15%NDA · Oct 31, 2006External genital and perianal wartsRed latex of the treeCroton lechleriCrofelemer, an oligomericproanthocyanidin mixtureFulyzaq, now Mytesi, 125 mgtabletNDA · Dec 31, 2012Non-infectious diarrhea in adults withHIV/AIDS on antiretroviral therapyPineapple stemAnanas comosusAnacaulase-bcdb, a mixture ofproteolytic enzymesNexoBridBLA · Dec 28, 2022Eschar removal in deeppartial-thickness and full-thicknessthermal burnsBirch barkBetula pendula, B.pubescensBirch triterpenes, a refined dryextractFilsuvez topical gel 10%NDA · Dec 18, 2023Wounds of dystrophic and junctionalepidermolysis bullosaFour approved botanical drugs with plant source, drug substance and approved useVeregen ointment 15% · NDA · Oct 31, 2006Green tea leaf (Camellia sinensis): Sinecatechins, apartially purified water-extract fraction. Use: Externalgenital and perianal wartsFulyzaq, now Mytesi, 125 mg tablet · NDA · Dec 31,2012Red latex of the tree (Croton lechleri): Crofelemer, anoligomeric proanthocyanidin mixture. Use:Non-infectious diarrhea in adults with HIV/AIDS onantiretroviral therapyNexoBrid · BLA · Dec 28, 2022Pineapple stem (Ananas comosus): Anacaulase-bcdb, amixture of proteolytic enzymes. Use: Eschar removal indeep partial-thickness and full-thickness thermal burnsFilsuvez topical gel 10% · NDA · Dec 18, 2023Birch bark (Betula pendula, B. pubescens): Birchtriterpenes, a refined dry extract. Use: Wounds ofdystrophic and junctional epidermolysis bullosa
Drugs@FDA (NDA 021902, NDA 202292, BLA 761192, NDA 215064) and the current labels, sections 1 and 11. Park et al. 2025 list these four as the approved botanical drugs.
Each approved botanical drug substance is specified as a mixture, down to constituents under 1%FDA labels
Range bars of label composition specifications in percent by weight on a log scale: Veregen catechins 85 to 95, EGCG more than 55; Filsuvez betulin 72 to 88, betulinic acid 2.6 to 4.2, lupeol 2.4 to 5.7, erythrodiol 0.5 to 1.2, oleanolic acid 0.3 to 0.8; NexoBrid listed proteins 80 to 95Veregen (sinecatechins), green tea leafCamellia sinensis· NDA 021902CatechinsCatechins: 85–95%85–95%Epigallocatechin gallateEpigallocatechin gallate: more than 55%more than 55%Filsuvez (birch triterpenes), birch barkBetula pendula, B. pubescens· NDA 215064BetulinBetulin: 72–88%72–88%Betulinic acidBetulinic acid: 2.6–4.2%2.6–4.2%LupeolLupeol: 2.4–5.7%2.4–5.7%ErythrodiolErythrodiol: 0.5–1.2%0.5–1.2%Oleanolic acidOleanolic acid: 0.3–0.8%0.3–0.8%NexoBrid (anacaulase-bcdb), pineapple stemAnanas comosus· BLA 761192Listed proteinsListed proteins: 80–95%80–95%0.1%1%10%100%Specified content, % by weight of the drug substance (logarithmic scale)Mytesi (crofelemer; NDA 202292) specifies an average degree of polymerization of 5 to 7.5 for its proanthocyanidin oligomers, a measure rather than a percentage.
Section 11 (Description) of the Veregen, Filsuvez, NexoBrid and Mytesi labels, retrieved from the openFDA label database, October 23, 2026.
Data: label specifications
  • Veregen: catechins 85 to 95% by weight of the drug substance, which “includes more than 55% of Epigallocatechin gallate (EGCg).”
  • Filsuvez: betulin 72–88%, lupeol 2.4–5.7%, betulinic acid 2.6–4.2%, erythrodiol 0.5–1.2%, oleanolic acid 0.3–0.8% (w/w).
  • NexoBrid: the listed proteins make up 80% to 95% w/w of anacaulase-bcdb.
  • Mytesi: average degree of polymerization of the oligomers 5 to 7.5.
Explanation, details and sources

Half of those INDs had been withdrawn or terminated by December 2020. Three of the four approved products are applied to skin or wounds; crofelemer is the only oral one. A disease claim needs this route (Course 10). An ingredient approved or substantially investigated as a drug first can also be excluded from the supplement definition, § 201(ff)(3)(B); see Legal status.

Barrier 5 · Ownership and access

A plant as found in nature cannot be patented, and CITES and access-and-benefit-sharing rules govern who may collect and trade it

CITES lists about 34,300 plant species, almost all in Appendix II; about 1,280 recorded medicinal species are among themCITES and Kew data
Log-scale bars of CITES-listed plant species: Appendix I about 395, Appendix II about 33,764, Appendix III about 134; about 1,280 of the 28,187 medicinal species are under CITES protectionAppendix ICommercial international trade prohibitedCITES Appendix I plants: 395 species, 4 subspecies, 12 varieties395 speciesAppendix IITrade authorized by export permitCITES Appendix II plants: 33,764 species including 110 populations33,764 speciesAppendix IIIListed at the request of a regulating PartyCITES Appendix III plants: 134 species and 1 variety134 speciesMedicinal species under CITESKew, 2017, of 28,187 medicinal speciesAbout 1,280 medicinal species are under CITES protectionabout 1,280101001,00010,000100,000Plant species (logarithmic scale)
CITES Secretariat, “The CITES species,” counts as of February 23, 2023 (34,293 plant species, plus 4 subspecies and 13 varieties, 34,310 taxa in all; Appendix II includes 110 populations); CITES, “The CITES Appendices”; Kew, State of the World's Plants 2017; CBD Secretariat, list of Parties and “About the Nagoya Protocol,” read October 23, 2026.
Explanation, details and sources

The Supreme Court held that “A naturally occurring DNA segment is a product of nature and not patent eligible merely because it has been isolated” (Myriad, 2013); Legal status covers what can be protected. The Nagoya Protocol entered into force on October 12, 2014 and has 143 Parties; the United States signed the Convention on Biological Diversity in 1993 but is not a Party.

For a U.S. firm the provider country's law still applies. Under the Protocol, utilization “includes research and development on the genetic or biochemical composition of genetic resources,” and access rules cover prior informed consent, mutually agreed terms and a permit or equivalent.

Barrier 6 · Variability

A natural mixture varies from batch to batch, so its effect has to be shown to be consistent

Four lines of evidence named by FDA support consistent effect across batchesFDA guidance
Four lines of evidence named by FDA, raw material control, chemical tests, manufacturing control, and biological assay with clinical data, together support consistent therapeutic effect across marketed batchesHeterogeneous mixture, uncertain active constituents“Totality of the evidence”Botanical drug substancea natural mixture whose batches vary with theplant, its growing and its processingBotanical raw material controle.g. agricultural practice and collectionQuality control by chemical testsspectroscopic or chromatographic methods thatcapture the active or chemical constituentsManufacturing controle.g. process validationBiological assay and clinical dataan assay that reflects the known or intendedmechanism; multiple-batch and dose-response clinicaldataConsistenttherapeutic effectacross marketedbatchesFour lines of evidence named by FDA support consistent therapeutic effect across batchesBotanical drug substancea natural mixture; batches varyBotanical raw material controle.g. agricultural practice and collectionQuality control by chemical testsspectroscopic or chromatographic methods thatcapture the active or chemical constituentsManufacturing controle.g. process validationBiological assay and clinical dataan assay that reflects the known or intendedmechanism; multiple-batch and dose-responseclinical dataConsistent therapeutic effectacross marketed batches
FDA, Botanical Drug Development, Guidance for Industry, Revision 1 (December 2016), Section III.
Explanation, details and sources

FDA's botanical drug guidance names, as “one of the critical issues,” ensuring “that the therapeutic effect for marketed drug product batches is consistent,” supported by a “totality of the evidence” approach. The same variability affects supplement specifications; Course 5 explains its sources.

Putting it together

Most barriers bind at more than one stage, so solving them one at a time rarely clears a candidate

Seven barriers across six development stagesEditorial assessment
Symbol matrix of seven barriers by six development stages; filled circles mark the stage where a barrier binds, half circles where it contributes, open circles where it is minorSourcingCharacterizationSafety studiesU.S. filingLabel andclaimsScale-up andsupplyIdentity and standardizationIdentity and standardization · Sourcing: highIdentity and standardization · Characterization: highIdentity and standardization · Safety studies: mediumIdentity and standardization · U.S. filing: highIdentity and standardization · Label and claims: lowIdentity and standardization · Scale-up and supply: mediumSafety evidence for the intended intakeSafety evidence for the intended intake · Characterization: lowSafety evidence for the intended intake · Safety studies: highSafety evidence for the intended intake · U.S. filing: highSafety evidence for the intended intake · Label and claims: lowSafety evidence for the intended intake · Scale-up and supply: lowNDI status and the notificationNDI status and the notification · Sourcing: lowNDI status and the notification · Characterization: mediumNDI status and the notification · Safety studies: mediumNDI status and the notification · U.S. filing: highNDI status and the notification · Scale-up and supply: mediumDrug-first exclusion, § 201(ff)(3)(B)Drug-first exclusion, § 201(ff)(3)(B) · U.S. filing: highDrug-first exclusion, § 201(ff)(3)(B) · Label and claims: mediumPatent eligibility of natural productsPatent eligibility of natural products · Characterization: mediumPatent eligibility of natural products · Label and claims: lowPatent eligibility of natural products · Scale-up and supply: mediumCITES and access-and-benefit-sharingCITES and access-and-benefit-sharing · Sourcing: highCITES and access-and-benefit-sharing · Characterization: mediumCITES and access-and-benefit-sharing · U.S. filing: lowCITES and access-and-benefit-sharing · Scale-up and supply: highBatch and supply variabilityBatch and supply variability · Sourcing: highBatch and supply variability · Characterization: highBatch and supply variability · Safety studies: mediumBatch and supply variability · U.S. filing: lowBatch and supply variability · Label and claims: lowBatch and supply variability · Scale-up and supply: highBinding at this stageContributesMinorNot a factorEditorial assessment; rationale in the text version
Assessment by the course author from the rules cited on this page; not a measurement.
Text version: why each barrier binds where it does
  • Identity binds at sourcing, characterization and filing (Latin binomial, 190.6(b)(2); identity test on each ingredient, 111.75).
  • Safety evidence binds where studies are run and where the notification states the basis for safety (190.6(b)(4)).
  • NDI status binds at filing; a process change can reopen it when FDA's draft view of “chemically altered” applies.
  • The drug-first exclusion binds at filing and marketing (§ 201(ff)(3)(B)).
  • Patent eligibility matters when a characterized form or process is to be protected (35 U.S.C. 101).
  • CITES and access rules bind at collection and at every shipment.
  • Variability binds wherever a lot is sourced, tested or scaled.
Explanation, details and sources

For a product decision this sets the order of work. Identity and access come first because every later document depends on them; marketing history and the intended claim then decide whether the route is a notification, no filing, or a drug program, and so how much safety and clinical evidence to buy.

Explorer · Seven gates

Walking one candidate through seven gates shows where its route forks and where it stops

The statutory clocks on the supplement and drug routesStatute and regulation
Two time axes in days. Supplement route: an NDIN is filed at least 75 days before first marketing, and a structure/function notification is due within 30 days after first marketing. Drug route: an IND goes into effect 30 days after FDA receives it, then clinical studies and an NDA follow, with no fixed clock to approvalSupplement routeAt least 75 days: no interstate marketing after the NDIN filing dateat least 75 days, no interstate saleStructure/function notification due no later than 30 days after first marketing30 daysNDIN filing date21 CFR 190.6(a), (c)First marketingday 0Notification deadline21 CFR 101.93(a)(1)The NDIN step applies only to an NDI outside the food-supply exception; FDA's reply is not an approval.Drug routeIND receivedday 0IND in effectday 30, 21 CFR 312.40(b)Phase 1 to 3 studiesno fixed durationNDA reviewsubstantial evidence, § 505(d)Approvalneeded to claim to treat adiseaseStatutory clocks: NDIN at least 75 days before marketing; structure/function notification within 30 days after; IND in effect 30 days after receipt; NDA approval with no fixed clockSupplement routeNDIN filedat least 75 days before first marketing (21 CFR190.6)First marketingday 0Structure/function notificationno later than 30 days after first marketing (21 CFR101.93(a)(1))Drug routeIND receivedday 0IND in effectday 30 (21 CFR 312.40(b))Phase 1 to 3 studiesno fixed durationNDA approvalsubstantial evidence (§ 505(d)); needed to claim totreat a disease
21 CFR 190.6(a), (c); 21 CFR 101.93(a)(1); 21 CFR 312.40(b); FD&C Act § 505(d).
Gates cleared7 of 7
Minimum wait before first sale75 days
RouteNDIN

Supplement route with a notification filed at least 75 days before marketing, then a structure/function notice within 30 days after.

Status of each gate for this candidateEducational logic
  1. Identity: cleared
  2. Access: cleared
  3. Drug-first clause: cleared
  4. Marketing history: new dietary ingredient
  5. Food-supply exception: does not apply; NDIN required
  6. Safety basis: cleared
  7. Claim: structure/function statement
Method, assumptions and limits
What it does
Applies the order of questions on this page to one candidate and reports the route, the gates cleared, and the shortest statutory wait between filing and first interstate sale.
Rules used
Identity: 21 CFR 190.6(b)(2) and 111.75(a)(1)(i). Drug-first: § 201(ff)(3)(B). New dietary ingredient: § 413(d). Food-supply exception: § 413(a)(1). Notification 75 days before marketing: § 413(a)(2), 21 CFR 190.6. Structure/function notice within 30 days after: 21 CFR 101.93(a)(1). Disease claim: 21 CFR 101.93(f). IND in effect 30 days after receipt: 21 CFR 312.40(b).
Assumptions
“Believed so, no records” is planned as a new dietary ingredient, because FDA keeps no authoritative list. The 25-year figure is FDA's recommended minimum when history of use alone is relied on, a recommendation rather than a rule. “Chemically altered” follows FDA draft guidance and is left to counsel.
Valid range
One botanical ingredient intended for a U.S. dietary supplement. It does not cover conventional foods, cosmetics, state law, sanctions or import rules.
What it does not establish
It is a teaching aid, not legal advice or a safety conclusion. Counsel decides legal status; qualified laboratories and experts generate the evidence.
Verification
Six answer sets are checked against an independent Python implementation of the same rules (tests/numerics/ref/academy_untapped-botanicals.py) and by unit tests (tests/unit/acclaim.test.js).
Explanation, details and sources

Three statutory clocks frame the routes: 75 days before marketing for an NDIN, 30 days after for a structure/function notice, and 30 days from receipt before an IND takes effect.

Common misreadings

Seven pairs that are easy to confuse when reading this evidence

  • Recorded medicinal use≠clinical proof of effect
  • Cited in a regulatory publication≠lawful to sell in the United States
  • An NDIN acknowledged without objection≠FDA approval or a finding of safety
  • FDA's public NDIN list≠every notification FDA received
  • FDA's preparation hours≠the cost of the studies behind a submission
  • A botanical IND≠an approved botanical drug
  • CITES Appendix II≠a trade ban
Key terms

Nine terms used in this course

Show the 9 key terms
Accepted species
The name a checklist treats as correct; synonyms point to it.
Dietary ingredient
A vitamin, mineral, botanical, amino acid, dietary substance, or their extracts and combinations (§ 201(ff)(1)).
New dietary ingredient
One not marketed in the United States before October 15, 1994.
NDIN
Premarket notification of a new dietary ingredient, filed at least 75 days ahead.
IND
Investigational new drug application, needed to study a product as a drug.
Botanical drug
A drug from plants, algae or macroscopic fungi, usually a mixture.
Drug-first exclusion
§ 201(ff)(3)(B): an article first approved or investigated as a drug is not a supplement.
Access and benefit-sharing
Consent, agreed terms and benefit-sharing for using genetic resources.
Capitalized cost
Spending carried forward to approval at a cost of capital.
Check your understanding

Five questions on the untapped territory

Kew's 4,478 species are “cited in a medicinal regulatory publication.” Are they authorized for sale in the United States?

No. The publications include pharmacopoeias and regulatory texts from many countries. U.S. status depends on the product category, the ingredient's marketing history and the claims.

A supplier says its root extract was “sold before 1994.” What does FDA expect?

FDA keeps no authoritative list, so the firm must document that a supplement containing the ingredient, or the ingredient for use in supplements, was marketed before October 15, 1994. Without records, plan for a notification.

FDA estimates 20 hours to prepare an NDIN. Why is that not the cost of the safety case?

FDA's estimate covers extracting and summarizing information already in the company's files. Toxicology studies, chemistry and history-of-use research come before that and are not counted.

Why does a botanical drug label specify constituents present at under 1%?

The drug substance is a mixture. Ranges for minor constituents help show that each batch matches the material tested, which supports consistent therapeutic effect.

A species is in CITES Appendix II. Can it be traded?

Yes, if the exporting country grants an export permit. Appendix II controls trade; Appendix I prohibits it unless the purpose of the import is not commercial.

Next: why U.S. botanicals cannot claim to treat disease

Course 10 reads the statute, FDA's own examples and the botanical drug pathway, with a claim sorter.

Go to Course 10Discuss a candidate