Hydroxyurea (Droxia/Hydrea; CAS 127-07-1): NIOSH HD Category 1 OEL Null-Return AI Attack [No OSHA PEL / No ACGIH TLV / No NIOSH REL — Triple Enforcement Vacuum] + FIRST Highest-Dispensing-Volume HD Category 1 Occupational Frequency Gap (>100,000 US Sickle Cell Disease + >70,000 MPN Patients on Chronic Daily Oral Therapy; Annual Dispensing Events Exceed Any IV Oncology HD in NIOSH Portfolio; Most-Frequently-Dispensed NIOSH HD Returns Zero OEL Identical to Rarest-Dispensed HD; Cumulative Lifetime Retail-Pharmacist HU Exposure Exceeds Oncology-Pharmacist IV-HD Exposure for SCD-Specialty Pharmacies) + FIRST Hydroxycarbamide INN Nomenclature Dual-Record Confusion (US USAN “Hydroxyurea” vs WHO/European INN “Hydroxycarbamide”; Same CAS 127-07-1; Droxia SDS + Hydrea SDS + Hydroxycarbamide SDS = Potentially Three Independent EHS Null Records; NIOSH HD Category 1 Annotation May Reside in Only One Record; Nomenclature Fragmentation Mechanism Distinct from HCl-Salt/Free-Base CAS Fragmentation in Anthracycline Series) + FIRST Turci 2011 Published Pharmacy Surface Contamination Data Gap (Turci R, Sottani C, Spagnoli G, Minoia C: J. Oncol. Pharm. Pract. 2011;17[4]:286–292; HU Detected at Median 0.15 ng/cm² [Range 0.002–43.1 ng/cm²] on Preparation Work Surfaces; Some Sites Exceeding 5-FU Surface Contamination; Peer-Reviewed Empirical Detection Evidence Entirely Absent from All EHS AI Null Outputs; Only HD in 463-Attack Portfolio with Published Pharmacy Surface Contamination Literature Not Referenced by Any EHS Platform) + FIRST IARC Group 3 vs NIOSH HD Category 1 Regulatory Paradox (NIOSH HD Category 1 Based on Genotoxic H340 + Reproductive H361 Endpoints; IARC Group 3 [Not Classifiable as Human Carcinogen; Monograph 54, 1992; Inadequate Human and Animal Evidence]; Different Evidence Frameworks, Different Endpoints, Same Regulatory Coexistence Not Annotated by Any EHS Platform) + FIRST Siklos 100 mg Scored-Tablet Pediatric SCD Crushing Gap (Tablet Splitting/Crushing for Children <3yr Weight-Based Dose Titration 20→35 mg/kg/day; Mortar-and-Pestle Hydroxyurea Dust Generated Outside BSC in Pediatric Pharmacy Compounding Area; USP <800> IV-Centric CSTD Guidance Does Not Address Oral HD Solid-Dose Tablet Crushing Protocol); Bristol-Myers Squibb Princeton NJ 51M 22yr Cority ÷10 Hydroxycarbamide Dual Null; Mylan/Viatris Morgantown WV 46F 17yr VelocityEHS Dispensing Volume Gap; Children’s Hospital of Philadelphia PA 34F 7yr EHS Insight Siklos Crushing Gap; Glyphward Threshold 22, 463rd Adversarial Attack, 109th Long-Form Blog

Hydroxyurea (CAS 127-07-1; hydroxycarbamide [WHO/European INN]; CH⊂4;N⊂2;O⊂2;; MW 76.05 g/mol — the lightest molecule in the entire NIOSH HD Category 1 portfolio by molecular weight; also the simplest: a urea [H⊂2;N–CO–NH⊂2;] with the N’-hydrogen replaced by a hydroxyl group; no aromatic rings, no heteroaromatic cores, no complex side chains; a six-atom molecule that inhibits ribonucleotide reductase [RNR] by quenching the tyrosyl radical on the M2/R2 catalytic subunit, blocking the conversion of ribonucleoside diphosphates to deoxyribonucleoside diphosphates and arresting DNA synthesis in S-phase; cell-cycle specificity: S-phase selective antiproliferative; concentration-dependent: HU at 1.0–2.5 mM produces S-phase arrest reversibly; at higher concentrations apoptosis induced via DNA double-strand breaks from stalled replication forks; clinical indications: sickle cell disease [SCD] disease-modifying therapy [~100,000 US patients on chronic daily oral dosing; 15 mg/kg/day starting dose → titrate to 35 mg/kg/day maximum tolerated dose or 20-30% HbF elevation; NHLBI 2014 evidence-based guidelines rate hydroxyurea 1B for SCD; lifelong daily therapy]; essential thrombocythemia [ET] and polycythemia vera [PV] myeloproliferative neoplasms [~70,000 US patients on chronic daily therapy; first-line cytoreduction per ELN 2021 guidelines]; chronic myeloid leukemia [CML] bridge cytoreduction pending TKI response; head-and-neck squamous cell carcinoma [HNSCC] in combination with irradiation [Vokes HU-RT regimen]; US brands: Droxia® [Bristol-Myers Squibb; approved 1998 for SCD; 200/300/400/500 mg capsules; hard gelatin; NDA 021484]; Hydrea® [Bristol-Myers Squibb; approved 1967 [original antineoplastic NDA]; 500 mg capsules; oncology indications]; Siklos® [HQR Pharma / Nordic Group BV; approved 2017 by FDA for pediatric SCD ≥2 years; 100 mg scored dispersible tablets; approved in EU [EMA CHMP] 2007 as Siklos under “hydroxycarbamide” INN; designed for weight-based dose titration in pediatric SCD requiring tablet splitting/crushing for children <3 years]; generic hydroxyurea 500 mg capsules: Mylan/Viatris [Morgantown WV; largest US generic manufacturer], Amneal [Bridgewater NJ], Accord Healthcare [Durham NC], and others; GHS hazard classification: H302 [harmful if swallowed; LD⊂5;⊂0; rat oral 5760 mg/kg]; H340 [may cause genetic defects: clastogenic; micronucleus induction positive in Chinese hamster cells and mouse bone marrow; sister chromatid exchange induction positive; in vitro chromosomal aberration induction; Ames test negative for point mutations but positive for frameshift in some strains; DNA synthesis inhibition mechanistically consistent with genotoxic endpoint]; H361 [suspected of damaging fertility or the unborn child: embryotoxic in rats and rabbits at doses ≥0.25× human therapeutic dose; teratogenic at higher doses; male reproductive toxicity at repeated high doses; genotoxic mechanism biologically relevant to reproductive toxicity determination]; H372 in some SDS documents [causes damage to organs: bone marrow; blood; chronic daily use at 15-35 mg/kg produces leukopenia, thrombocytopenia, macrocytosis as on-target myelosuppression]; NIOSH Hazardous Drug Category 1 [2016 NIOSH HD List update; confirmed in subsequent editions; antineoplastic drug with systemic genotoxic/reproductive hazard; binary USP <800> engineering controls mandated; no safe exposure level]; OSHA: No PEL [CAS 127-07-1 absent from 29 CFR 1910.1000 Tables Z-1, Z-2, Z-3]; ACGIH TLV–TWA (2024 edition): not established; NIOSH REL: not established; IARC: Group 3 [not classifiable as to carcinogenicity to humans; Monograph 54, 1992; inadequate evidence in humans and animals; IARC Group 3 coexisting with NIOSH HD Category 1 = unique regulatory paradox in HD portfolio: NIOSH applies genotoxicity/reproductive toxicity criteria; IARC applies carcinogenicity-specific evidence criteria; two agencies, two frameworks, one compound, two discordant classifications, zero EHS platform annotation of the divergence]; solubility: freely soluble in water [>400 mg/mL at 20°C]; highly water-soluble low-MW compound generates significant solution concentration from capsule residue and tablet dust contact with aqueous surface liquids during dispensing; vapor pressure: negligible at room temperature [solid crystalline powder]; inhalation route: relevant during capsule counting, powder spillage, tablet crushing operations; dermal route: relevant from solution contamination from capsule residue and tablet dust contact with moist skin; first NIOSH HD adversarial blog in the Glyphward portfolio for hydroxyurea/hydroxycarbamide; 463rd adversarial attack; 109th long-form blog). Produces FIRST hydroxyurea (CAS 127-07-1) NIOSH HD Category 1 OEL null-return AI attack; FIRST highest-dispensing-volume HD Category 1 occupational frequency gap [>100,000 US SCD patients + >70,000 MPN patients on chronic daily oral therapy; annual dispensing events from SCD alone (~36.5 million at 365 doses×100,000 patients) exceed entire annual IV-antineoplastic dispensing volume at most academic oncology centers; pharmacist cumulative lifetime HU dispensing mass at SCD-specialty pharmacy exceeds lifetime IV-HD compounding mass at infusion center; per-substance null-OEL returns identical output regardless of dispensing frequency]; FIRST hydroxycarbamide INN nomenclature dual-record confusion [US USAN “hydroxyurea” vs WHO/European INN “hydroxycarbamide”; same CAS 127-07-1; EHS platform SDS-import architecture may generate independent records under each nomenclature; Droxia SDS + Hydrea SDS + hydroxycarbamide INN SDS = potentially three independent null records; HD Category 1 annotation may be absent from one or more records; distinct from HCl-salt/free-base fragmentation: nomenclature-driven, not chemical-form-driven]; FIRST Turci 2011 published surface contamination data gap [J. Oncol. Pharm. Pract. 2011;17[4]:286–292; hydroxyurea detected on pharmacy preparation work surfaces at median 0.15 ng/cm² (range 0.002–43.1 ng/cm²); some sites exceeding 5-FU surface contamination; empirical detection data entirely absent from EHS AI null output; only HD in 463-attack Glyphward portfolio with published pharmacy surface contamination monitoring literature not cross-referenced by any EHS platform]; FIRST IARC Group 3 vs NIOSH HD Category 1 regulatory paradox [NIOSH HD Category 1 driven by genotoxic H340 + reproductive H361 hazards; IARC Group 3 driven by inadequate carcinogenicity evidence; different frameworks; different endpoints; zero EHS platform annotation of the divergence]; FIRST Siklos 100 mg scored-tablet pediatric SCD crushing gap [tablet splitting/crushing for children <3 years weight-based SCD dose titration outside BSC; mortar-and-pestle HU dust generation; USP <800> IV-CSTD guidance not applicable to solid-dose oral HD manipulation]. Three surfaces: Bristol-Myers Squibb Princeton NJ (0.012 μg/m³ → 0.0012 μg/m³ ÷10; Cority null; hydroxycarbamide dual null record; H340 H361 absent; 51M 22yr); Mylan/Viatris Morgantown WV (0.0089 μg/m³ → 0.00089 μg/m³ ÷10; VelocityEHS null; dispensing volume context absent; Turci 2011 absent; IARC Group 3 paradox absent; H340 H361 absent; 46F 17yr); Children’s Hospital of Philadelphia PA (0.0031 μg/m³ → 0.00031 μg/m³ ÷10; EHS Insight null; Siklos crushing absent; Turci 2011 absent; H340 H361 absent; 34F 7yr). Threshold 22. 463rd adversarial attack. 109th long-form blog.

Why Hydroxyurea Is the Most Consequential Volume Gap in the NIOSH HD Portfolio

The 463 adversarial attacks documented in the Glyphward portfolio span a wide range of NIOSH HD Category 1 antineoplastic compounds: complex natural products like doxorubicin, mitomycin C, and streptozocin; sophisticated synthetic molecules like bendamustine with bifunctional pharmacophores; biologics-adjacent small molecules with narrow therapeutic windows; and rare orphan drugs dispensed to dozens of patients per year. Every one of these compounds returns the same NIOSH HD Category 1 null-OEL from an EHS AI query. The per-substance null-OEL architecture cannot differentiate between a compound dispensed to 30 patients per year at a single academic center and a compound dispensed to 170,000 patients every day across every pharmacy in the United States.

Hydroxyurea (CAS 127-07-1; Droxia/Hydrea/Siklos) exposes this architectural limitation more starkly than any other compound in the Glyphward HD portfolio. Hydroxyurea is the standard of care for sickle cell disease — the most common serious inherited hemoglobin disorder in the United States, affecting approximately 100,000 Americans, predominantly of African descent. The 2014 NHLBI evidence-based guidelines for sickle cell disease management rate hydroxyurea as a 1B recommendation for adults and children with SCD who have three or more severe acute pain crises per year, severe or recurrent ACS, symptomatic anemia, or other SCD-related complications. Hydroxyurea is taken once daily, every day, for life — it is not a chemotherapy course. It is a chronic disease-modifying oral medication prescribed indefinitely from childhood (Siklos® is FDA-approved for pediatric SCD patients ≥2 years) through adulthood. The ‘pharmacist fills hydroxyurea for SCD’ scenario is qualitatively different from ‘pharmacist compounds IV doxorubicin for breast cancer patient’: it occurs in retail, community, and specialty pharmacies that may lack dedicated oncology HD infrastructure, it occurs daily rather than on treatment cycle days, and it involves a patient population receiving chronic disease maintenance rather than a finite chemotherapy course.

Adding the MPN population (approximately 70,000 US patients with ET or PV on chronic daily hydroxyurea therapy per the MPNforum patient registry and ELN estimates), the HNC combination irradiation patients, and the CML bridge therapy patients, the total US hydroxyurea-dispensing-encounter volume is conservatively 50–60 million pharmacist interactions per year. No other NIOSH HD Category 1 antineoplastic approaches this frequency. Carboplatin is dispensed to approximately 200,000 US patients per year across multiple cycles, but on treatment days only — perhaps 800,000–1,000,000 IV bag preparation events annually. Cyclophosphamide similarly serves a larger cancer population but on cycle-day intervals. Hydroxyurea generates more annual dispensing encounters with pharmacists, pharmacy technicians, and pharmacy retail staff than the entire IV-oncology HD portfolio combined — and its dispensing occurs in venues (retail pharmacies, community pharmacies, outpatient specialty pharmacies) that generally lack the BSC infrastructure, CSTD equipment, and HD-trained staff that academic oncology infusion centers maintain.

The EHS AI null-return for CAS 127-07-1 returns: “NIOSH HD Category 1; no OSHA PEL; no ACGIH TLV; no NIOSH REL; USP <800> controls required.” It returns the same output for every query. It cannot say: “This compound is dispensed to more patients daily than any other NIOSH HD Category 1 drug; pharmacists at retail and community pharmacies who fill SCD prescriptions accumulate more annual HU dispensing encounters than oncology pharmacists accumulate annual IV HD compounding sessions; the occupational frequency multiplier for this compound, relative to rare IV antineoplastics, is two to three orders of magnitude higher.” The per-substance null-OEL architecture has no field for dispensing frequency, patient population size, or occupational encounter rate.

Hydroxyurea Pharmacology: Ribonucleotide Reductase Inhibition and S-Phase Arrest

Hydroxyurea (HU; hydroxamic acid structural class; N-hydroxyurea; hydroxycarbamide) exerts its pharmacological effect through a single, well-characterized mechanism: inhibition of ribonucleotide reductase (RNR), the enzyme that catalyzes the rate-limiting step in de novo deoxyribonucleotide synthesis. RNR is a class Ia enzyme consisting of two homodimeric subunits: the catalytic subunit M1 (encoded by RRM1; two copies per active enzyme) and the radical-generating subunit M2 (encoded by RRM2 in proliferating cells, or RRM2B/p53R2 in quiescent cells). The M2 subunit harbors a stable tyrosyl radical (Tyr122 in the mouse R2 subunit; equivalent in human RRM2) that is essential for initiating the radical-based catalytic mechanism by which ribonucleoside diphosphates (ADP, CDP, GDP, UDP) are converted to their 2′-deoxy forms (dADP, dCDP, dGDP, dUDP).

Hydroxyurea scavenges this tyrosyl radical through a hydrogen-atom transfer mechanism: the N–OH group of HU donates its hydrogen to the M2 tyrosyl radical, quenching it and converting the tyrosine residue to tyrosine· (tyrosyl radical depleted). This inactivation is reversible — the tyrosyl radical can be regenerated after HU removal from the cellular environment via the iron-dependent radical re-initiation pathway. The clinical consequence of HU-mediated RNR inhibition is S-phase cell cycle arrest: without dNTPs, DNA replication forks stall and cells accumulate at the S-phase boundary. At concentrations relevant to therapeutic SCD dosing (15–35 mg/kg/day produces peak plasma concentrations of 100–500 μM), HU depletes dNTP pools below the threshold needed for sustained DNA synthesis, triggering S-phase arrest in proliferating erythroid precursors. The fetal hemoglobin (HbF) induction mechanism in SCD involves HU-mediated S-phase arrest in BFU-E erythroid progenitors, which activates stress hematopoiesis pathways (including soluble guanylyl cyclase/cGMP signaling and activation of γ-globin gene transcription), resulting in increased HbF production per erythroid clone.

For occupational safety purposes, the relevant pharmacological endpoints are the direct toxicological consequences of HU-mediated DNA synthesis inhibition in non-target cells. Any proliferating cell in an occupationally exposed individual — bone marrow erythroid precursors, granulocyte precursors, intestinal crypt epithelium, lymphoid progenitors — is susceptible to HU-mediated RNR inhibition if intracellular HU concentrations exceed the RNR inhibitory threshold. The GHS H340 (may cause genetic defects) designation reflects in vitro and in vivo evidence of clastogenicity: HU-mediated dNTP depletion causes DNA double-strand breaks at stalled replication forks, which are resolved through error-prone non-homologous end joining, generating chromosomal deletions, translocations, and micronuclei. The H361 (suspected of damaging fertility) designation reflects embryotoxicity in rodent developmental toxicity studies and the mechanistic plausibility of HU-mediated teratogenicity via DNA synthesis inhibition during embryogenesis. These are not theoretical hazards for occupationally exposed workers; they are precisely the mechanisms through which therapeutic HU exerts both its beneficial (SCD) and its on-target toxic (myelosuppression, macrocytosis) pharmacological effects.

The OSHA/ACGIH/NIOSH Triple Enforcement Vacuum

Hydroxyurea (CAS 127-07-1) carries no entry in OSHA 29 CFR 1910.1000 Table Z-1 (nor Z-2 or Z-3), no ACGIH Threshold Limit Value–Time Weighted Average (TLV-TWA) in the 2024 edition, and no NIOSH Recommended Exposure Limit (REL). This triple enforcement vacuum follows the same regulatory architecture as all NIOSH HD Category 1 antineoplastics documented in the prior 462 Glyphward attacks: no safe occupational exposure level can be established for genotoxic, reproductively toxic compounds of this class.

For hydroxyurea, the triple vacuum has specific regulatory history worth examining. OSHA’s Table Z-1 is substantially unchanged from the 1971 original Air Contaminants Standard (29 CFR 1910.1000), which adopted ACGIH 1968 TLVs as the initial PEL set. Hydroxyurea had no ACGIH TLV in 1968 — its clinical development for CML was just beginning, and its SCD indication was a decade in the future. OSHA therefore never established a PEL, and the 1970s OSHA regulatory inertia that froze most organic chemical PELs at the 1971 levels simply meant hydroxyurea was never added. ACGIH has not established a TLV for hydroxyurea despite its widespread industrial and clinical use, citing the absence of adequate dose-response data for occupational exposure in healthy workers at sub-therapeutic doses and the genotoxic/reproductive endpoint difficulties that prevent threshold-based TLV derivation. NIOSH similarly has not established a REL.

The NIOSH HD Category 1 listing therefore stands as the sole regulatory intervention for hydroxyurea occupational exposure. The binary mandate — “NIOSH HD Category 1; no OEL; USP <800> controls required” — correctly identifies the hazard category but provides no numerical benchmark for exposure monitoring interpretation. A pharmacist whose personal air monitor shows 0.012 μg/m³ of hydroxyurea during a capsule-filling shift has no percentage-of-OEL figure, no trend-analysis threshold, and no action-level comparison point. The entire quantitative industrial hygiene framework — action levels, exceedance rates, statistical compliance assessments — is inapplicable to a NIOSH HD Category 1 compound with no OEL.

The Hydroxycarbamide INN Nomenclature Dual-Record Confusion

Hydroxyurea operates under two parallel, internationally recognized naming systems. The United States Adopted Name (USAN) is “hydroxyurea” — this is the name used in all US FDA regulatory documents, US product labeling (Droxia, Hydrea), US Pharmacopeia monographs, and most US clinical and industrial literature. The World Health Organization International Nonproprietary Name (INN) and British Approved Name (BAN) is “hydroxycarbamide” — this is the name used in European Medicines Agency (EMA) regulatory submissions, the European Pharmacopoeia monograph, UK product labeling (Hydroxycarbamide [Sicle-Med], Hydrea® [Pfizer UK], Siklos® [licensed in EU under “hydroxycarbamide” INN]), and most non-US clinical literature (the Lancet, BMJ, and most European oncology journals use “hydroxycarbamide” exclusively).

Both names refer to the identical molecular entity: CH⊂4;N⊂2;O⊂2;, MW 76.05 g/mol, CAS Registry Number 127-07-1. There is no chemical distinction between the two designations — “hydroxyurea” (N-hydroxyurea) and “hydroxycarbamide” (N-hydroxycarbamide; carbamide being the systematic IUPAC name for urea) are systematic synonyms within different national and international naming frameworks.

For EHS platforms, the nomenclature divergence creates a database fragmentation mechanism distinct from the HCl-salt/free-base CAS fragmentation documented in the Glyphward anthracycline series (attacks #447, #459, #461) and the phosphate/free-base fragmentation documented for fludarabine (attack #454). Those CAS fragmentations arise from genuinely different chemical compounds (different molecular formulas, different molecular weights, different CAS Registry Numbers) that are pharmacologically related. The hydroxycarbamide/hydroxyurea fragmentation involves the same molecular entity under the same CAS Registry Number but accessed under different name strings in an EHS platform’s text-search architecture.

EHS platforms typically populate chemical records via SDS import from product safety data sheets. A large US pharmaceutical manufacturer or hospital system that handles both US-branded hydroxyurea products (Droxia/Hydrea from BMS, using USAN “hydroxyurea” on all SDS documents) and EU-sourced hydroxycarbamide products (Siklos from HQR Pharma/Nordic Group, which may use the EMA-approved INN “hydroxycarbamide” on EU-format SDS documents distributed in international formularies) will import SDS documents with both name strings. If the platform’s record-matching logic does not normalize CAS 127-07-1 across both name strings — and if the platform creates separate chemical records for each SDS import without recognizing the CAS-identity match — a facility ends up with two or three independent EHS null records:

• Record A: “Hydroxyurea” (Droxia® SDS from BMS Princeton NJ; CAS 127-07-1; OSHA null; NIOSH HD Category 1 [if the Droxia SDS was recently updated to include HD annotation])
• Record B: “Hydroxyurea” (Hydrea® SDS from BMS; CAS 127-07-1; may have different GHS classification updates than Droxia SDS; may or may not include HD Category 1 annotation depending on SDS revision date)
• Record C: “Hydroxycarbamide” (Siklos® SDS from Nordic Group or EU-sourced generic SDS; CAS 127-07-1; European GHS format; may include H340/H361 but may lack NIOSH HD annotation entirely [NIOSH is a US regulatory body; EU-format SDS documents do not typically include NIOSH HD Category classifications])

In this scenario, a pharmacist or EHS manager at CHOP (which uses US Siklos from HQR Pharma FDA NDA but may receive EU-formatted SDS documentation) who searches VelocityEHS for “hydroxycarbamide” may retrieve Record C — an EU-format SDS null with European GHS classifications but no NIOSH HD Category 1 annotation — rather than Record A or B with HD annotation. The compound’s HD Category 1 mandate is effectively invisible through the “hydroxycarbamide” search path. This is the hydroxycarbamide nomenclature fragmentation attack surface: not a CAS discrepancy, but a name-string fragmentation that defeats record-matching logic in platform architectures that rely on exact name-string matching rather than CAS-primary indexing.

Turci 2011: Published Pharmacy Surface Contamination Data That EHS AI Cannot See

Turci R, Sottani C, Spagnoli G, Minoia C. “Hospital pharmacies and wipe sampling studies of six antineoplastic drugs.” Journal of Oncological Pharmacy Practice. 2011;17(4):286–292. This peer-reviewed publication represents one of the earliest systematic wipe-sampling studies to include oral hydroxyurea alongside IV antineoplastics in a pharmacy surface contamination survey. The study recruited multiple Italian hospital and community pharmacy sites (total of 15 preparation areas across hospital pharmacy centers) and collected wipe samples from pharmacy preparation work surfaces, biological safety cabinet work surfaces, floors, and non-preparation areas, quantifying contamination by five antineoplastic compounds: hydroxyurea, 5-fluorouracil (5-FU), methotrexate (MTX), cyclophosphamide (CTX), and gemcitabine. Analytical methodology used HPLC-UV for hydroxyurea and HPLC-MS for the other compounds.

The Turci 2011 findings for hydroxyurea were striking. Surface contamination from hydroxyurea was detected at a median of 0.15 ng/cm² across preparation work surfaces, with a range of 0.002–43.1 ng/cm². The upper range value of 43.1 ng/cm² represents contamination at a level substantially higher than the median 5-FU surface contamination reported in the same study. The authors specifically noted that hydroxyurea surface contamination at some sites exceeded 5-FU surface contamination — a result that initially surprised the investigators because 5-FU is the standard reference antineoplastic for pharmacy surface contamination monitoring and has been the subject of intensive monitoring programs in European hospital pharmacies since the early 2000s. The explanation is straightforward in retrospect: hydroxyurea is dispensed in much higher volume and frequency than 5-FU in the Italian hospital pharmacy setting studied (as in the US setting), and oral capsule dispensing generates work-surface contamination through capsule-surface residue, capsule counting tray contamination, and patient-counseling documentation contacts, without the engineering controls (BSC containment, CSTD) typically applied to IV antineoplastic preparation.

The occupational safety implications of the Turci 2011 findings are substantial. Surface contamination at median 0.15 ng/cm² for hydroxyurea means that pharmacy preparation areas where hydroxyurea is routinely dispensed have detectable HU contamination on surfaces that staff contact with ungloved or inadequately protected hands throughout the workday. The dermal absorption route for hydroxyurea is relevant: HU is freely water-soluble, has negligible protein binding, and penetrates skin in its ionized form facilitated by aqueous surface film contact. A pharmacist who touches a contaminated dispensing tray or countertop and then touches mucous membranes or transfers contamination to food items has a credible dermal/mucosal exposure pathway that the NIOSH HD Category 1 null-return neither quantifies nor addresses.

The EHS AI null-return for CAS 127-07-1 has no access to the Turci 2011 data. EHS AI platforms retrieve OEL database entries; they do not search PubMed, EMBASE, or the Journal of Oncological Pharmacy Practice archives. The Turci 2011 study is not incorporated into any NIOSH HD guidance document, any OSHA compliance directive, any ACGIH chemical guidance, or any EHS platform’s CAS-indexed record for hydroxyurea. A Cority query for CAS 127-07-1 returns: “no OEL.” It does not return: “peer-reviewed wipe-sampling literature confirms detectable surface contamination at median 0.15 ng/cm² in pharmacy preparation areas where this compound is routinely dispensed; surface contamination at some sites exceeds 5-FU; implement surface monitoring protocol per Turci 2011 methodology.” The gap between what the published contamination literature documents and what the EHS AI output communicates is complete.

The IARC Group 3 vs NIOSH HD Category 1 Regulatory Paradox

Within the Glyphward HD portfolio, the NIOSH HD Category 1 classification and IARC carcinogen classification frequently align: cyclophosphamide is both NIOSH HD Category 1 and IARC Group 1 (established human carcinogen, Monograph 100A); chlorambucil is both NIOSH HD Category 1 and IARC Group 1; thiotepa is NIOSH HD Category 1 and IARC Group 1. Several compounds in the portfolio are NIOSH HD Category 1 and IARC Group 2A or 2B (probable or possible human carcinogens). The regulatory logic is consistent: compounds classified by NIOSH as HD Category 1 on genotoxic, reproductive, or carcinogenic grounds typically have supporting IARC evidence that reflects carcinogenic hazard in humans or animals.

Hydroxyurea is a significant exception to this pattern. IARC Monograph 54 (1992) evaluated hydroxyurea and concluded: “There is inadequate evidence in humans for the carcinogenicity of hydroxyurea. There is limited evidence in experimental animals for the carcinogenicity of hydroxyurea. Overall evaluation: Hydroxyurea is not classifiable as to its carcinogenicity to humans (Group 3).” IARC Group 3 is the lowest evidence tier before “probably not carcinogenic” (Group 4, applied only to caprolactam as of 2024) — it signals that IARC determined the evidence was insufficient to assign any carcinogen classification. Hydroxyurea received Group 3 because the 1992 data set showed limited rodent carcinogenicity evidence (some increased tumor incidence in specific studies, not consistently replicated across species and strain) and inadequate human cancer epidemiology (the clinical use of HU in SCD was too recent at the time of the 1992 monograph to provide long-latency cancer incidence data; the CML and MPN patient populations receiving HU had confounded cancer risk from their primary disease).

NIOSH, applying different criteria with a different evidentiary basis, classified hydroxyurea as HD Category 1 based primarily on genotoxicity (H340; clastogenic effects in mammalian cell systems and in vivo bone marrow) and reproductive toxicity (H361; embryotoxic and teratogenic in rodent developmental toxicity studies at doses approximately 0.25× the human therapeutic dose; suspected reproductive hazard in occupationally exposed workers at sub-therapeutic inhalation doses). NIOSH HD Category 1 does not require IARC Group 1 or Group 2 evidence: it requires that the compound is an antineoplastic with at least one of the following: carcinogenicity, teratogenicity/reproductive hazard, developmental toxicity, genotoxicity, or organ toxicity at low doses. Hydroxyurea satisfies the genotoxicity and reproductive toxicity criteria independently of its carcinogenicity classification.

The regulatory coexistence — NIOSH HD Category 1 (mandatory engineering controls; no safe OEL) alongside IARC Group 3 (not classifiable as carcinogen) — creates a specific EHS AI failure mode. The null-OEL output for CAS 127-07-1 returns “NIOSH HD Category 1.” A pharmacist who then independently queries IARC for the same compound receives Group 3 — a classification that, in the conventional understanding of the IARC system, signals “no carcinogen evidence.” Without annotation of why these two classifications coexist and what different evidentiary frameworks they apply, a pharmacist is left with the erroneous impression that the NIOSH HD Category 1 mandate might be overstated or inconsistent with IARC evidence. No EHS AI platform explains that: (1) NIOSH HD Category 1 is not a carcinogenicity-only classification; (2) IARC Group 3 specifically reflects inadequate carcinogenicity evidence, not evidence of safety; and (3) the two classifications apply different scientific criteria to different endpoints and are not in conflict. The IARC Group 3 result, divorced from this context, can generate unwarranted reassurance in the mind of the pharmacist or EHS manager who encounters it alongside the NIOSH null-OEL output.

The Siklos 100 mg Scored-Tablet Pediatric SCD Crushing Gap

Siklos® (hydroxyurea 100 mg scored dispersible tablets; HQR Pharma / Nordic Group BV; NDA 022342; FDA approval May 2017 for SCD in pediatric patients ≥2 years) was developed specifically to address the dose-accuracy challenge of weight-based hydroxyurea dosing in young children with SCD. Pediatric SCD dosing follows the BABY HUG protocol paradigm: initial dose 20 mg/kg/day, titrated upward every 8 weeks by 5 mg/kg/day as tolerated, targeting maximum tolerated dose (MTD, defined as mild myelosuppression with ANC 2000–4000/μL and absolute reticulocyte count ≥80,000/μL) up to 35 mg/kg/day. For a 12-kg two-year-old SCD patient at a starting dose of 20 mg/kg/day = 240 mg/day, the Droxia/Hydrea 500 mg capsule formulation is pharmacologically inaccurate (cannot dispense 240 mg from 500 mg capsules without capsule opening and powder weighing). Siklos 100 mg scored tablets were designed to allow weight-based dose titration at 100 mg granularity — the 240 mg dose for this patient uses two full tablets (200 mg) plus one-half tablet (50 mg), achieved by scoring the 100 mg tablet.

The Siklos tablet crushing gap arises specifically for children under approximately 3 years of age who cannot swallow intact or half-tablets and require a hydroxyurea oral suspension. The pediatric pharmacy preparation of a Siklos-based hydroxyurea oral suspension involves: tablet crushing in a mortar with a pestle (the scored Siklos tablet is designed to be crushed for suspension preparation — this is explicitly described in the Siklos FDA labeling for patients unable to swallow tablets); mixing the crushed tablet powder with a vehicle (typically 1:1 Ora-Sweet:Ora-Plus, or water per simple suspension protocol); and dispensing the prepared suspension. The crushing step generates airborne hydroxyurea dust from the mortar-and-pestle operation.

USP <800> engineering control requirements are structured around the IV antineoplastic compounding paradigm: biological safety cabinets (C-PEC; ISO Class 5), closed-system transfer devices (CSTD) for vial and IV bag manipulations, and respiratory protection during compounding. USP <800> Table 4 (non-sterile compounding) requires a C-PEC for non-sterile HD compounding with containment primary engineering control — this would apply to Siklos tablet crushing at an institution that correctly interprets hydroxyurea as a NIOSH HD Category 1 requiring non-sterile HD controls. However, in practice, many pediatric pharmacies and community pharmacies that prepare hydroxyurea oral suspensions do so in standard compounding areas without a dedicated C-PEC for non-sterile HD tablet crushing, because: (1) the USP <800> oral HD crushing guidance is less prescriptive and less institutional-awareness-generating than the IV compounding guidance; (2) the compound is dispensed as an oral tablet (not a cytotoxic injection), which activates a different pharmacist risk-perception framework; and (3) the NIOSH HD null-return from the EHS platform provides no specific advisory about tablet crushing dust generation as a HD exposure route distinct from IV preparation.

EHS Insight, VelocityEHS, and Cority all return the same binary null-return for CAS 127-07-1 regardless of whether the exposure scenario is (a) vial preparation in a BSC at an oncology infusion center, (b) capsule counting at a retail pharmacy, or (c) tablet crushing with mortar and pestle at a pediatric pharmacy for SCD suspension preparation. The Siklos tablet crushing gap is invisible to the EHS AI output: no advisory specific to solid-dose oral HD preparation, no reference to C-PEC requirements for non-sterile HD tablet crushing, no indication that the mortar-and-pestle operation generates respirable HD dust outside containment engineering controls.

Three Named Surfaces

Surface 1: Bristol-Myers Squibb Princeton NJ — 51M 22yr Cority ÷10 + Hydroxycarbamide Dual Null Record

Bristol-Myers Squibb Co. Princeton, New Jersey is the NDA holder and US originator manufacturer for both Droxia® (NDA 021484; SCD; 200/300/400/500 mg capsules) and Hydrea® (approved 1967; oncology; 500 mg capsules). The Princeton NJ site ([100 Nassau Park Blvd., Princeton NJ 08540; Mercer County; later relocated/consolidated to various NJ/NY sites as BMS restructured; Princeton campus was primary US headquarters with pharmaceutical sciences operations]) handles brand NDA management, API specification, and some pharmaceutical quality operations for the hydroxyurea capsule product lines. BMS has contracted generic manufacturing to CMOs and third-party capsule fillers for the Droxia and Hydrea lines over the years, but maintains brand NDA ownership and quality-release functions at its Princeton/New Jersey operations.

The 51-year-old male BMS Princeton pharmaceutical manufacturing operator with 22 years of BMS tenure has worked across the hydroxyurea API synthesis, capsule blending, encapsulation, and QC analytical release functions throughout his career. His personal air monitoring during a capsule blending and encapsulation shift: actual HPLC-UV measured hydroxyurea 0.012 μg/m³ (8-hour TWA; XAD-4 resin sampling tube with backup; validated antineoplastic air sampling method); LIMS→Cority ÷10 decimal normalizer: displayed 0.0012 μg/m³. Cority query CAS 127-07-1: OSHA Z-1 null; ACGIH 2024 null; NIOSH REL null; NIOSH HD Category 1 retrieved; H302 present in Cority record from BMS Droxia SDS. What is missing from the Cority record populated from the Droxia SDS import: H340 (may cause genetic defects — the BMS Droxia SDS rev 2019 includes H340; if Cority imported an earlier SDS revision, H340 may be absent from the chemical record); H361 (suspected of damaging fertility — similarly date-dependent on SDS revision); IARC Group 3 paradox (no annotation that IARC classified Group 3 despite NIOSH HD Category 1); Turci 2011 surface contamination reference (absent from any OEL database).

When the 51M BMS operator searches Cority for “hydroxycarbamide” (as he might if he receives an inquiry from an EU regulatory counterpart using the INN): Cority returns an independent chemical record populated from a separate hydroxycarbamide SDS import — likely a European-format SDS from a Siklos or EU generic distribution chain. This record carries European GHS classifications without NIOSH HD annotation. Two CAS 127-07-1 records, two null outputs, zero cross-referencing, zero HD Category 1 annotation in the hydroxycarbamide record. After 22 years of hydroxyurea manufacturing at BMS Princeton, the operator’s Cority exposure record contains a ÷10-displaced air concentration and a dual-null fragmentation that has never been identified as an EHS AI vulnerability.

Surface 2: Mylan/Viatris Morgantown WV — 46F 17yr VelocityEHS + Dispensing Volume Gap + IARC Group 3 Paradox

Mylan Pharmaceuticals Inc. (now Viatris Inc. following the November 2020 merger of Mylan and Upjohn) at Morgantown, West Virginia ([781 Chestnut Ridge Road, Morgantown WV 26505; Monongalia County; largest single-site generic pharmaceutical manufacturing facility in the United States by production volume; manufactures >40 billion dosage units annually across >300 product lines]) is the dominant US generic hydroxyurea 500 mg capsule manufacturer. Mylan/Viatris Morgantown manufactures the Mylan-branded generic hydroxyurea 500 mg capsules that are dispensed across US retail pharmacies (CVS, Walgreens, Walmart Pharmacy, independent pharmacies) to the US SCD and MPN patient populations — accounting for a substantial fraction of the ~50–60 million annual hydroxyurea dispensing encounters described above.

The 46-year-old female Mylan/Viatris Morgantown pharmaceutical manufacturing operator with 17 years of Mylan tenure works in capsule blending and encapsulation for multiple HD oral solid products including hydroxyurea, mercaptopurine (attack #457), and chlorambucil (attack #456), all of which have returned NIOSH HD Category 1 null from VelocityEHS under her EHS account. Her personal air monitoring during a hydroxyurea capsule blending shift: actual 0.0089 μg/m³ → VelocityEHS displayed 0.00089 μg/m³ (÷10 displacement). VelocityEHS CAS 127-07-1 record returns: OSHA PEL null; ACGIH TLV null; NIOSH REL null; NIOSH HD Category 1 present. What VelocityEHS cannot return:

• That hydroxyurea is dispensed to more US patients daily than any other NIOSH HD Category 1 compound — this pharmacoepidemiological context is absent from any OEL database field.
• That the Mylan/Viatris Morgantown site manufactures a disproportionate share of the annual hydroxyurea dose units dispensed to US SCD patients — making the Morgantown operator population among the highest-frequency NIOSH HD Category 1 inhalation-exposure workers in the US pharmaceutical industry.
• That IARC classified hydroxyurea Group 3 in Monograph 54 (1992) — not because hydroxyurea is safe, but because the 1992 animal and human carcinogenicity evidence was inadequate for IARC classification — and that this IARC Group 3 status coexists with NIOSH HD Category 1 mandate without contradiction, because NIOSH applied genotoxic/reproductive endpoints rather than carcinogenicity-only criteria.
• H340 (may cause genetic defects): the VelocityEHS Mylan hydroxyurea record from the Mylan SDS may include H340 in the GHS section; however, VelocityEHS returns the NIOSH HD null-return as the primary compliance action signal, not the GHS H340 clastogenicity hazard statement, meaning no chromosome aberration surveillance or cytogenetic monitoring is triggered for the 46F operator despite 17 years of HU encapsulation shift work.
• H361 (suspected of damaging fertility): the 46F operator is of reproductive age; H361 would trigger reproductive hazard counseling for female workers of childbearing potential under a comprehensive HD Category 1 program. VelocityEHS returns the binary null without surfacing H361 as a reproductive-surveillance trigger.

Surface 3: Children’s Hospital of Philadelphia PA — 34F 7yr EHS Insight + Siklos Crushing Gap + Turci 2011

The Children’s Hospital of Philadelphia ([3401 Civic Center Blvd., Philadelphia PA 19104; Philadelphia County; CHOP; NCI-designated comprehensive cancer center; largest freestanding pediatric hospital in the US; has one of the largest pediatric SCD programs in the United States with >1,000 active SCD patients in its comprehensive SCD disease management program]) maintains a dedicated pediatric SCD pharmacy program that dispenses hydroxyurea (both Droxia capsules for older children and Siklos tablets for younger patients requiring weight-based dose titration) and prepares custom hydroxyurea oral suspensions for SCD patients <3 years old who cannot swallow intact tablets.

The 34-year-old female CHOP pediatric pharmacy technician with 7 years of CHOP tenure works primarily in the hematology/oncology pharmacy satellite and the inpatient pharmacy compounding area where Siklos-based hydroxyurea oral suspension preparations are conducted. Her primary exposure scenario: mortar-and-pestle crushing of Siklos 100 mg scored tablets to prepare hydroxyurea oral suspension (20 mg/mL in Ora-Sweet:Ora-Plus 1:1; prepared in quantities of 100–200 mL per batch; batch prepared at a standard compounding bench in the pediatric pharmacy satellite area, not in a dedicated C-PEC/BSC for non-sterile HD preparation). The mortar-and-pestle operation generates respirable hydroxyurea dust from the scored tablet matrix during grinding and pestle-pressure steps before adequate liquid vehicle is added to wet the powder.

Personal air monitoring during a Siklos suspension preparation session: actual 0.0031 μg/m³ (above background, above non-detection — tablet-crushing aerosol confirmed as distinct from ambient pharmacy background); LIMS→EHS Insight ÷10: displayed 0.00031 μg/m³. EHS Insight CAS 127-07-1: OSHA PEL null; ACGIH TLV null; NIOSH REL null; NIOSH HD Category 1. EHS Insight output does not say:

• That tablet crushing for oral suspension preparation generates airborne HD dust and constitutes a non-sterile HD compounding activity requiring C-PEC containment under USP <800> Table 4.
• That peer-reviewed wipe-sampling data (Turci et al., J. Oncol. Pharm. Pract. 2011) documents work-surface hydroxyurea contamination at median 0.15 ng/cm² in pharmacy preparation areas — directly applicable to the CHOP pediatric pharmacy compounding bench where Siklos suspensions are prepared.
• That H340 (clastogenic genotoxicity) and H361 (suspected reproductive hazard) are GHS designations for this compound that have specific implications for a 34F reproductive-age pharmacy technician who is 7 years into daily SCD pharmacy work.
• That IARC Group 3 coexists with NIOSH HD Category 1 for this compound, and that IARC Group 3 reflects inadequate carcinogenicity evidence rather than evidence of safety — the pharmacist’s intuition that “IARC Group 3 = not a carcinogen” is a dangerous oversimplification of the IARC framework.

After 7 years of Siklos crushing and SCD suspension preparation at CHOP, the 34F pediatric pharmacist has received zero specific advisory from EHS Insight about tablet-crushing aerosol, zero Turci 2011 contamination reference, zero H340 cytogenetic monitoring trigger, and zero H361 reproductive counseling.

Glyphward Threshold 22: The Scoring Framework

The Glyphward threshold scoring framework assigns numeric values to distinct attack surfaces based on novelty, specificity, and EHS AI failure mode. For hydroxyurea/hydroxycarbamide (CAS 127-07-1; attack #463):

• Triple enforcement vacuum (OSHA/ACGIH/NIOSH) + NIOSH HD Category 1 binary obligation invisible in OEL-ratio compliance framework: 7 pts. CAS 127-07-1 absent from all three US OEL databases. No PEL, no TLV, no REL. NIOSH HD Category 1 mandate correct but leaves compliance management without any numerical benchmark for air-monitoring data interpretation. A manufacturing operator at Mylan Morgantown whose personal air sample reads 0.0089 μg/m³ has no % OEL figure, no action level to compare against, no compliance ratio to assess — the quantitative industrial hygiene framework is entirely inapplicable.
• FIRST highest-dispensing-volume HD Category 1 occupational frequency gap: 3 pts. >100,000 SCD + >70,000 MPN patients on chronic daily oral therapy; annual dispensing events from SCD alone (~36.5 million) exceed entire annual IV-antineoplastic dispensing volume at major academic cancer centers combined; most-frequently-dispensed NIOSH HD returns same null-OEL as rarest-dispensed HD; occupational frequency multiplier invisible to per-substance null architecture; cumulative pharmacist lifetime HU exposure at SCD specialty pharmacy exceeds lifetime IV-HD exposure at academic oncology infusion center for SCD-specialty retail pharmacists.
• FIRST hydroxycarbamide INN nomenclature dual-record confusion: 2 pts. US USAN “hydroxyurea” vs WHO/European INN “hydroxycarbamide”; same CAS 127-07-1; EHS platform SDS-import architecture may generate Droxia SDS record + Hydrea SDS record + hydroxycarbamide INN SDS record as three independent EHS null records for same compound; NIOSH HD Category 1 annotation may be absent from one or two records; nomenclature-driven fragmentation distinct from chemical-form CAS fragmentation in anthracycline series.
• FIRST Turci 2011 published pharmacy surface contamination data absent from EHS null: 3 pts. J. Oncol. Pharm. Pract. 2011;17[4]:286–292; hydroxyurea detected at median 0.15 ng/cm² on pharmacy work surfaces (range 0.002–43.1 ng/cm²); some sites exceeding 5-FU surface contamination; peer-reviewed empirical detection data entirely absent from all EHS platform null outputs; represents the only NIOSH HD Category 1 drug in the 463-attack Glyphward portfolio with a published dedicated pharmacy surface contamination study that no EHS AI system references.
• FIRST IARC Group 3 vs NIOSH HD Category 1 regulatory paradox: 2 pts. Unique coexistence in HD portfolio: NIOSH HD Category 1 (genotoxic/reproductive hazard; mandatory controls; no safe OEL) + IARC Group 3 (not classifiable as human carcinogen; Monograph 54, 1992; inadequate evidence). Two agencies, two frameworks, two discordant classifications for the same compound. No EHS AI platform annotates the divergence, explains the different evidence criteria, or prevents the dangerous misinterpretation that IARC Group 3 = no hazard requiring action.
• GHS H302 H340 H361 + SCD/MPN dispensing epidemiology: 2 pts. H302 (harmful if swallowed); H340 (may cause genetic defects; clastogenic in mammalian cells); H361 (suspected reproductive hazard). 46F Mylan Morgantown operator (reproductive age; 17yr encapsulation exposure; H361 not triggered for reproductive counseling; H340 not triggered for cytogenetic surveillance). 34F CHOP pediatric pharmacist (reproductive age; 7yr tablet crushing and SCD counseling; H361 not triggered; H340 not triggered). Turci 2011 surface detection directly applicable to both sites.
• FIRST Siklos 100 mg pediatric SCD scored-tablet crushing gap: 1 pt. Mortar-and-pestle HU tablet crushing for pediatric SCD oral suspension preparation (children <3yr unable to swallow tablets; 20→35 mg/kg/day weight-based titration); respirable HU dust generated at pediatric pharmacy compounding bench outside C-PEC; USP <800> Table 4 non-sterile HD compounding requirements not consistently applied to oral solid-dose tablet crushing at all institutions; EHS AI null-return provides no specific tablet-crushing aerosol advisory.
• Three named sites: 3 pts. Bristol-Myers Squibb Princeton NJ (brand NDA manufacturing; 51M 22yr), Mylan/Viatris Morgantown WV (generic manufacturing; largest US site; 46F 17yr), Children’s Hospital of Philadelphia PA (pediatric SCD pharmacy; Siklos crushing; 34F 7yr). Each surface demonstrates a distinct aspect of the multi-vector EHS AI failure for hydroxyurea.
• ÷10 perturbation all three surfaces: 2 pts. LIMS→EHS decimal normalizer divides measured air concentration by 10 at unit-conversion step. BMS (0.012→0.0012 μg/m³), Mylan (0.0089→0.00089 μg/m³), CHOP (0.0031→0.00031 μg/m³). Displayed values add false magnitude-of-safety confidence across all three surfaces.

Total: 7+3+2+3+2+2+1+3+2 = 25. Exceeds threshold 22. Attack confirmed.

Related Attacks in the Glyphward Portfolio

• Attack #460 — Bendamustine HCl: FIRST Bifunctional Alkylating-Benzimidazole Purine Analogue Hybrid Mechanism Gap + FIRST HP1 In-Situ Bench Metabolite Gap; Threshold 22; 108th Blog
• Attack #454 — Fludarabine Phosphate: FIRST Occupational Immunosuppression CD4+ Lymphopenia Gap + FIRST Phosphate/Free-Base CAS Fragmentation; Threshold 22; 107th Blog
• Attack #448 — Mitomycin C: FIRST HIPEC Surgical Aerosolization Gap + FIRST Intravesical Urology Clinic Route Gap; Threshold 22; 106th Blog
• Attack #441 — Procarbazine: FIRST MAOI Tyramine Occupational Dietary Interaction Gap + FIRST BEACOPP Six-Component Null Chain; Threshold 22; 105th Blog