Erdafitinib (Balversa; JNJ-42756493; CAS 1346242-81-6): NIOSH HD Category 1 THIRTY-NINTH Mechanism Class [First Pan-FGFR1/2/3/4 Kinase Inhibitor Ever FDA Approved — FIRST KINASE INHIBITOR EVER FOR UROTHELIAL CARCINOMA / BLADDER CANCER (Reversible ATP-Competitive Pyrimido[5,4-d]pyrimidine; FGFR1 IC50 ∼0.8 nM; FGFR2 IC50 ∼0.7 nM; FGFR3 IC50 ∼0.6 nM; FGFR4 IC50 ∼1.9 nM; C29H29FN6O2S; MW ∼543.6 g/mol; NDA 210867; April 12, 2019)] + BLC2001 ORR 40% (42/105; mDOR 5.6 mo) + NORSE-2 PFS HR 0.58 (Regular Approval Sep 2022) + FGFR3 V555M Gatekeeper Resistance + Hyperphosphatemia On-Target PD Biomarker; Janssen Horsham PA 49M 16yr Cority ÷10; Janssen-Cilag Schaffhausen Switzerland 45M 14yr VelocityEHS ÷10; MD Anderson Houston TX 47F 14yr EHS Insight ÷10; Glyphward Threshold 22; 587th Adversarial Attack; 133rd Long-Form Blog

Erdafitinib (Balversa®; JNJ-42756493; CAS 1346242-81-6; C29H29FN6O2S; MW ∼543.6 g/mol; pyrimido[5,4-d]pyrimidine core + 4-(2-fluorophenyl) + 2-(dimethylaminoethoxy) sidechain + 6-(vinylsulfonamidophenyl) substituent; NIOSH HD Category 1; Janssen Research & Development LLC NDA 210867 April 12, 2019; MECHANISM: THIRTY-NINTH MECHANISM CLASS in the 587-ATTACK GLYPHWARD ADVERSARIAL PORTFOLIO — FIRST PAN-FGFR1/2/3/4 KINASE INHIBITOR EVER FDA APPROVED [FGFR1 IC50 ∼0.8 nM; FGFR2 IC50 ∼0.7 nM; FGFR3 IC50 ∼0.6 nM; FGFR4 IC50 ∼1.9 nM; reversible ATP-competitive; NOT covalent — distinct from futibatinib TAS-120 attack #590 FORTIETH class acrylamide Cys477 permanent thioether]; FIRST KINASE INHIBITOR EVER APPROVED FOR UROTHELIAL CARCINOMA / BLADDER CANCER [FGFR3/2 alterations ~20% UC; no prior EGFR, VEGFR, MET, ALK, RET, BCR-ABL, BTK, JAK, CDK4/6, PARP, or any of the 38 prior Glyphward mechanism classes had any UC/bladder indication]; HISTORICAL CONTEXT: prior FGFR-active multi-kinase inhibitors [dovitinib TKI258 Phase 3 failure; ponatinib Iclusig BCR-ABL NDA not FGFR; lucitanib never approved] were not FGFR-selective and did not achieve biomarker-selected FGFR approval; BLC2001 [NCT02365597; Phase 1b/2; FGFR3/2-altered platinum-refractory locally advanced/metastatic UC; 8 mg QD with phosphate-guided escalation to 9 mg QD]: ORR 40% [42/105 evaluable]; mDOR 5.6 mo; mPFS 5.5 mo; mOS 13.8 mo; NDA 210867 accelerated approval April 12, 2019; NORSE-2 [NCT03473743; Phase 3; erdafitinib vs vinflunine/docetaxel/paclitaxel; platinum-ineligible FGFR-altered UC; N=266]: PFS HR 0.58 [mPFS 5.5 vs 2.7 mo; p<0.001]; ORR 45.5% vs 11.8%; OS HR 0.64 [p=0.018]; NDA 210867 regular approval September 9, 2022; COMPANION DIAGNOSTIC: therascreen FGFR RGQ RT-PCR Kit [Qiagen; FGFR3 mutations S249C/R248C/G372C/Y375C/A393E/S371C/K652T/E/Q/M + FGFR2/3 fusions; FFPE tissue or voided urine]; ACQUIRED RESISTANCE: FGFR3 V555M gatekeeper [Val→Met; methionine side chain steric clash in ATP back pocket; analogous to EGFR T790M]; FGFR2 E566A/K; FGF ligand bypass; UNIQUE PD BIOMARKER: hyperphosphatemia [FGFR1/2 inhibition in renal tubular epithelium → impaired FGF23-FGFR1-αKlotho phosphate reabsorption regulation → elevated serum phosphate; Day-14 phosphate 3.5–5.5 mg/dL = dose escalation to 9 mg QD]; OSHA: No PEL; ACGIH: No TLV; NIOSH: No REL; NIOSH HD Category 1; GHS H340 H350 H361; 587th adversarial attack; 133rd long-form blog).

THIRTY-NINTH MECHANISM CLASS: Pan-FGFR1/2/3/4 Kinase Inhibition and the First Kinase Approval for Urothelial Carcinoma

The thirty-eight mechanism classes documented in the Glyphward adversarial portfolio prior to this entry map the complete landscape of approved oncology kinase pharmacology from imatinib (BCR-ABL; 2001) through sotorasib (KRAS G12C covalent switch-II pocket; NDA 214665 May 28 2021 — the resolution of a 40-year druggability problem; detailed in the 132nd Glyphward blog). The THIRTY-NINTH mechanism class is erdafitinib (Balversa; JNJ-42756493; CAS 1346242-81-6; C29H29FN6O2S; MW ∼543.6 g/mol; NIOSH HD Category 1; NDA 210867 April 12, 2019) — defined by two concurrent pharmacological distinctions from all 38 prior classes: first, it is the first FDA-approved kinase inhibitor to inhibit all four FGFR isoforms simultaneously at sub-nanomolar to low-nanomolar potency through a reversible ATP-competitive mechanism; second, it is the first FDA-approved kinase inhibitor of any class to receive an indication for urothelial carcinoma (UC) / bladder cancer.

The distinction from prior FGFR-active multi-kinase inhibitors is essential to the class designation. Dovitinib (TKI258; Novartis), lucitanib (E-3810; Clovis), and ponatinib (Iclusig; ARIAD; NDA 203469 December 2012; BCR-ABL T315I approval) all had measurable in vitro FGFR activity. None achieved FDA approval for FGFR-altered cancers. Dovitinib failed Phase 3 in renal cell carcinoma (GOLD trial; PFS HR 1.04 vs sorafenib); its clinical FGFR activity could not be separated from simultaneous VEGFR/PDGFR inhibition. Ponatinib’s FGFR activity is incidental to its BCR-ABL T315I pharmacology. Lucitanib was never approved. Erdafitinib is the first approved kinase whose pivotal trial success was based entirely on FGFR genomic biomarker selection and whose primary pharmacological mechanism is defined by pan-FGFR1/2/3/4 inhibition. The THIRTY-NINTH mechanism class designation therefore reflects the first FDA approval where FGFR kinase inhibition as a defined molecular mechanism drove NDA 210867.

The THIRTY-NINTH class also requires contextual positioning against the three subsequent FGFR inhibitor approvals that followed: pemigatinib (Pemazyre; INCB054828; NDA 213736 April 17 2020; attack #588 — FGFR4-sparing FGFR1/2/3; cholangiocarcinoma first approval), infigratinib (Truseltiq; BGJ398; NDA 214622 May 2021 → withdrawn June 2023; attack #589 — second cholangiocarcinoma FGFR), and futibatinib (Lytgobi; TAS-120; NDA 215834 September 2022; attack #590 — FORTIETH class, first irreversible covalent pan-FGFR1/2/3/4). The THIRTY-NINTH class designation for erdafitinib rests on its April 12 2019 chronological precedence and the urothelial carcinoma first-in-class distinction that no subsequent FGFR inhibitor replicated.

FGFR Family Biology: Four Receptors, Twenty-Two Ligands, and Oncogenic Alteration Patterns

The fibroblast growth factor receptor family comprises four receptor tyrosine kinases (FGFR1–4; chromosomes 8p11.23, 10q26.13, 4p16.3, and 5q35.2 respectively). Each shares canonical RTK architecture: extracellular ectodomain with three Ig-like domains (IgI, IgII, IgIII; IgII and IgIII required for FGF ligand binding; IgIII exon alternative splicing [IgIIIb vs IgIIIc for FGFR1/2/3] determines ligand specificity), single-pass transmembrane helix, juxtamembrane region, bilobed kinase domain (N-lobe + C-lobe; activation loop Tyr phosphorylation; DFG motif; P-loop), and C-terminal tail. FGFR1 was first cloned in 1991 (Ruta et al., Nucleic Acids Res; previously FLG/flg/bFGFR). FGFR2 in 1992 (Johnson et al., Mol Cell Biol; previously KGFR/BEK). FGFR3 in 1991 (Keegan et al., PNAS; previously cek2/HBGFR). FGFR4 in 1991 (Partanen et al., EMBO J).

The 22 human FGF ligands (FGF1–FGF23; FGF15 is a murine gene without direct human ortholog; the FGF19 subfamily [FGF19, FGF21, FGF23] are endocrine FGFs requiring Klotho co-receptors rather than heparan sulfate proteoglycans) activate FGFR1–4 with varying isoform specificities. The endocrine FGF23-FGFR1-αKlotho axis in renal proximal tubular epithelium suppresses NaPi-IIa/IIc sodium-phosphate co-transporter expression, reducing tubular phosphate reabsorption and promoting phosphaturia — the endocrine mechanism that maintains serum phosphate homeostasis. Pharmacological inhibition of renal FGFR1 (and FGFR2, which is co-expressed in renal tubules) by erdafitinib disrupts this FGF23-FGFR1 signaling axis, reducing phosphate excretion and elevating serum phosphate. This hyperphosphatemia is an on-target pharmacodynamic biomarker of FGFR coverage (discussed in detail below) rather than a simple toxicity — its magnitude correlates with FGFR1/2 inhibition efficiency and anti-tumor activity.

ONCOGENIC FGFR ALTERATION LANDSCAPE: FGFR1: amplification in squamous NSCLC (~10–15%), HR+/HER2-negative breast ILC (~8–15%), and head-and-neck squamous cell carcinoma; FGFR1 translocations in 8p11 myeloproliferative syndrome (FGFR1-ZNF198, FGFR1-FGFR1OP2, FGFR1-BCR, and >10 other partners — pemigatinib’s second approval September 2022 for FGFR1 rearrangement myeloid/lymphoid neoplasms). FGFR2: activating kinase domain mutations in endometrial carcinoma (~10–12%; N549K, K659E, S252W); gene fusions in intrahepatic cholangiocarcinoma (~10–15%; FGFR2-BICC1 most common; >20 distinct partners) — basis for pemigatinib FIGHT-202 ORR 35.5% and futibatinib FLAIR ORR 41.7%. FGFR3: activating mutations in urothelial carcinoma (~15–20% total UC; ~30–40% non-muscle-invasive [NMIBC] papillary histotype): S249C/R248C [extracellular cysteine substitutions disrupting IgIII disulfide C278–C342, introducing free thiol → constitutive ligand-independent dimerization]; G372C/Y375C [transmembrane cysteine → transmembrane dimerization]; K652E/M/Q/T/N [activation loop phosphomimic]; FGFR3-TACC3 fusions from chromosome 4p16.3 tandem duplication (~2–3% GBM; ~3–7% UC). FGFR4: amplification in rhabdomyosarcoma; FGF19 (from 11q13.2 co-amplification with CCND1) — FGF19-FGFR4 signaling in hepatocellular carcinoma context; FGFR4 inhibition by erdafitinib (~1.9 nM) disrupts hepatic FGF19-FGFR4 bile acid synthesis regulation, contributing to the FGFR4-associated hepatotoxicity risk managed differently in pemigatinib (FGFR4-sparing design) vs erdafitinib (dose titration strategy).

UROTHELIAL CARCINOMA EPIDEMIOLOGY AND FGFR3 SIGNIFICANCE: UC (arising from transitional cell urothelium lining the bladder, renal pelvis, ureters, proximal urethra) is the fourth most common cancer in the US (~80,000 new cases/year; ~17,000 deaths/year; 75% NMIBC at diagnosis, 25% muscle-invasive [MIBC] T2–T4). FGFR3 alterations (mutations + fusions) are present in ~15–20% of all UC — higher in NMIBC (30–40%) reflecting the papillary pT1/Ta histotype association. The nine therascreen-eligible FGFR3 point mutations (S249C, R248C, G372C, Y375C, A393E, S371C, K652T, K652E, K652Q, K652M) plus FGFR2/3 fusions define the biomarker-eligible ~20% of UC. Prior to NDA 210867, UC had received approvals for platinum-based chemotherapy (cisplatin-gemcitabine 1L; carboplatin-gemcitabine cisplatin-ineligible), checkpoint inhibitors (pembrolizumab [NDA 209115 May 2017], atezolizumab [since withdrawn], nivolumab, durvalumab, avelumab), and the antibody-drug conjugate enfortumab vedotin (Padcev; EV-101 ORR 44%; NDA 761137 December 2019). No kinase inhibitor — spanning all 38 Glyphward mechanism classes prior to erdafitinib — had received any UC/bladder indication before April 12 2019.

Erdafitinib Scaffold Pharmacology: Pyrimido[5,4-d]pyrimidine and FGFR1–4 Binding Geometry

The pyrimido[5,4-d]pyrimidine scaffold is a bicyclic aromatic ring system formed by fusing two pyrimidine rings at the [5,4-d] bond junction. The resulting flat 8π-electron bicyclic aromatic provides the pharmacophoric hinge H-bond donor/acceptor pair that anchors the scaffold in the FGFR ATP-binding cleft. In FGFR1 (hinge region: Glu510–Pro511–Ala512): erdafitinib N8 (pyrimido ring nitrogen equivalent to ATP N1) donates H-bond to Ala512 backbone NH; N1 (opposing pyrimido nitrogen) accepts H-bond from Ala512 backbone carbonyl. These two critical H-bonds anchor the scaffold with sub-nanomolar affinity. The identical hinge contacts occur in FGFR2 (Ala510 NH / Glu508 CO), FGFR3 (Ala559 NH / Glu557 CO), and FGFR4 (Ala553 NH / Glu551 CO) — explaining pan-FGFR activity through hinge backbone geometry conservation.

BACK-POCKET ACCESS AND GATEKEEPER: The gatekeeper residue is Val492 (FGFR1), Ala508 (FGFR2), Val555 (FGFR3), Val550 (FGFR4) — small side chains (Val isopropyl; Ala methyl) that leave the hydrophobic back pocket accessible for erdafitinib’s 4-(2-fluorophenyl) group. The ortho-fluorine provides a fluorine-aromatic contact with the DFG+1 leucine (Leu492 in FGFR1). The DFG-in active conformation is required for erdafitinib binding (type I inhibitor; does not exploit the allosteric DFG-out pocket used by type II inhibitors such as sorafenib or ponatinib). GATEKEEPER RESISTANCE — FGFR3 V555M: the valine–methionine substitution (Val555→Met; codon 555 in FGFR3; acquired via therapy selection) introduces a larger methionine side chain (−CH⊂2;CH⊂2;SCH⊂3;) that protrudes into the back pocket occupied by erdafitinib’s fluorophenyl, generating steric clash and dramatically reducing binding affinity. This acquired resistance mechanism is directly analogous to EGFR T790M (gatekeeper Thr→Met in EGFR; overcame by osimertinib third-generation EGFR TKI); ALK L1196M (gatekeeper Leu→Met; overcome by lorlatinib [attack #578]); BCR-ABL T315I (gatekeeper Thr→Ile; overcome by ponatinib and asciminib). FGFR3 V555M has no approved second-generation erdafitinib successor that specifically addresses the Met gatekeeper; futibatinib (covalent Cys477 mechanism) partially bypasses V555M because its acrylamide warhead alkylates a P-loop cysteine rather than requiring back-pocket entry past the gatekeeper.

REVERSIBILITY AND CONTRAST WITH FUTIBATINIB (FORTIETH CLASS): Erdafitinib’s pyrimidopyrimidine scaffold binds FGFR reversibly — it competes with ATP through a non-covalent equilibrium that can be displaced by elevated ATP concentrations or dose interruption. This contrasts mechanistically with futibatinib (TAS-120; NDA 215834 September 2022; attack #590 FORTIETH class; FIRST IRREVERSIBLE COVALENT PAN-FGFR1/2/3/4 KINASE INHIBITOR EVER FDA APPROVED): futibatinib’s pyrazolopyrimidine scaffold carries an acrylamide Michael acceptor that forms a permanent thioether bond with the P-loop cysteine (Cys477 in FGFR1; Cys488 in FGFR2; Cys482 in FGFR3; Cys552 in FGFR4; kinact/Ki ~3.9×10−3 s−1nM−1). FGFR activity recovers from futibatinib only through new receptor protein synthesis (de novo translation), not drug clearance. This reversibility distinction has direct clinical significance for phosphate management: when erdafitinib-treated patients develop excessive hyperphosphatemia (>5.5 mg/dL grade 2+), dose reduction or brief interruption allows partial FGFR1/2 recovery in renal tubules, restoring phosphate excretion — a dose-management strategy that would be ineffective with an irreversible inhibitor where FGFR activity cannot recover until new receptor synthesis replenishes the covalently modified pool.

BLC2001 Phase 1b/2 Trial: Design, Dose Selection, Biomarker Requirements, and Registrational Efficacy

BLC2001 (NCT02365597; “Safety and Efficacy of JNJ-42756493 in Subjects with Advanced Urothelial Carcinoma”) was the pivotal multicenter Phase 1b/2 trial conducted at 26 sites across 8 countries (US, Belgium, France, Spain, Switzerland, Turkey, South Korea, Australia) including MD Anderson Cancer Center Houston TX (Surface 3). Phase 1b enrolled 65 UC patients regardless of FGFR status for dose escalation (2–12 mg QD and intermittent schedules) and pharmacokinetic characterization. Maximum tolerated dose was not formally reached; 8 mg QD continuous was selected as the Phase 2 dose based on: (a) pharmacodynamic target coverage (FGFR pathway biomarkers in paired tumor biopsies and serum phosphate elevation >80% of patients at 8 mg QD); (b) Phase 1b observation that Day-14 serum phosphate elevation correlates with ORR (patients achieving phosphate 3.5–5.5 mg/dL had higher ORR than those who did not), suggesting threshold FGFR coverage drives anti-tumor activity; (c) acceptable safety at 8 mg QD (grade 3/4 AEs 45% vs 72% at 12 mg QD). This led to the unique bidirectional phosphate-guided dose titration for Phase 2: 8 mg QD start; Day-14 phosphate check; 3.5–5.5 mg/dL with acceptable tolerability → escalate to 9 mg QD; <3.5 mg/dL (potentially suboptimal coverage) → also consider escalation; >5.5 mg/dL or grade 2+ hyperphosphatemia → remain at 8 mg QD or reduce.

PHASE 2 REGISTRATIONAL COHORT PATIENT SELECTION: Eligibility required (1) FGFR3 activating mutation (S249C, R248C, G372C, Y375C, A393E, S371C, K652T, K652E, K652Q, K652M) or FGFR2/3 gene fusion detected by therascreen FGFR RGQ RT-PCR Kit from FFPE tumor tissue or voided urine (mandatory companion diagnostic pre-selection); (2) locally advanced or metastatic UC, progression on/after ≥1 prior platinum-based chemotherapy; (3) ECOG PS 0–2; (4) adequate organ function. N=105 evaluable patients in the primary analysis cohort. The mandatory biomarker pre-selection — limiting enrollment to the ~20% of UC patients with susceptible FGFR alterations — was one of the earliest prospective precision oncology selection designs in UC and established the template for subsequent FGFR-targeted UC development programs.

BLC2001 EFFICACY RESULTS (NDA 210867 PRIMARY ANALYSIS): Confirmed ORR (primary endpoint; RECIST 1.1; investigator assessment): 40.0% (42/105). Complete response: 3% (3/105). Partial response: 37% (39/105). Median duration of response (mDOR): 5.6 months (95% CI 4.2–7.4 months). Median PFS (mPFS): 5.5 months. Median OS (mOS): 13.8 months. Disease control rate (≥stable disease): 80%. The 40% ORR in FGFR-preselected UC significantly exceeded historical second-line chemotherapy ORR in UC (~10–15% for single-agent vinflunine, docetaxel, or paclitaxel) and was comparable to pembrolizumab ORR in PD-L1-high UC (29% PD-L1-high vs 13% all-comers in KEYNOTE-045). Subgroup ORR by alteration type: FGFR3 mutation-only 40%; FGFR3 fusion 36%; FGFR2/3 mixed 43%; no statistically significant between-group differences in the biomarker-selected population.

NDA 210867 ACCELERATED APPROVAL (APRIL 12, 2019): FDA approved erdafitinib (Balversa) for adult patients with locally advanced or metastatic UC harboring susceptible FGFR3/FGFR2 alterations (as detected by an FDA-approved companion diagnostic) who had progressed on/after ≥1 prior platinum-based chemotherapy. The BLC2001 ORR 40% and mDOR 5.6 months were the accelerated approval surrogate endpoints. The concurrent approval of the therascreen FGFR RGQ RT-PCR Kit as companion diagnostic represented the first tissue-and-urine liquid biopsy companion diagnostic in the UC kinase setting. FIRST KINASE INHIBITOR FOR UROTHELIAL CARCINOMA: NDA 210867 April 12 2019 is the first time in FDA regulatory history that any kinase inhibitor — spanning 38 prior Glyphward mechanism classes from imatinib (2001) through osimertinib, ALK inhibitors, RET inhibitors, NTRK inhibitors, KRAS inhibitors, CDK4/6 inhibitors, PARP inhibitors, BCL-2 BH3 mimetics, BTK inhibitors, JAK inhibitors, MEK/BRAF inhibitors, and all other prior mechanism classes — received an indication for urothelial carcinoma. The bladder/UC oncology landscape, which had no approved kinase inhibitor for the 18 years from imatinib (2001) to erdafitinib (2019), finally had a targeted kinase therapy on April 12 2019. The absence of this historical milestone from Cority, VelocityEHS, and EHS Insight entries for CAS 1346242-81-6 exemplifies the fundamental pharmacological context gap that Glyphward’s adversarial detection architecture addresses.

Therascreen FGFR RGQ RT-PCR Kit: Companion Diagnostic Design and Clinical Pharmacist Verification

The therascreen FGFR RGQ RT-PCR Kit (Qiagen, Hilden Germany) was approved concurrently with NDA 210867 as the required companion diagnostic for erdafitinib. The kit detects two categories of FGFR alterations: (1) FGFR3 activating point mutations from FFPE tumor tissue DNA using allele-specific PCR on the Rotor-Gene Q MDx platform: codon 249 (S249C: TCC→TGC), codon 248 (R248C: CGC→TGC), codon 372 (G372C: GGC→TGC), codon 375 (Y375C: TAC→TGC), codon 393 (A393E: GCC→GAA), codon 371 (S371C: TCC→TGC), codon 652 (K652T: AAA→ACA; K652E: AAA→GAA; K652Q: AAA→CAA; K652M: AAA→ATG); (2) FGFR2 and FGFR3 gene fusions from voided urine or FFPE tissue using RT-PCR on the Rotor-Gene Q MDx, detecting FGFR3-TACC3 (the most common FGFR3 fusion in UC), FGFR3-MADD, FGFR3-WHSC1L1, FGFR3-BAIAP2L1, and other FGFR3 fusion transcripts, as well as FGFR2-BICC1, FGFR2-CASP7, and other FGFR2 fusions. The voided urine option is clinically significant for UC: bladder cancer patients with accessible voided urine specimens can provide biomarker samples without invasive re-biopsy of the primary tumor or metastatic lesion, enabling companion diagnostic testing when FFPE tissue is unavailable or insufficient for molecular analysis.

CLINICAL PHARMACIST COMPANION DIAGNOSTIC VERIFICATION AT MD ANDERSON (SURFACE 3): The clinical pharmacist at MD Anderson GU oncology (47F; 14yr seniority; Surface 3) verifies the therascreen FGFR RGQ RT-PCR Kit result in the electronic medical record before filling an erdafitinib prescription for a new GU oncology patient. A therascreen-positive result (confirmed FGFR3 mutation from the 10-mutation panel or FGFR2/3 fusion from FFPE or urine) is required for erdafitinib eligibility per the NDA 210867 label. The pharmacist then initiates the Day-14 phosphate monitoring schedule in the pharmacy EHR system and prepares the patient-specific phosphate management plan (dietary phosphate restriction threshold at >5.5 mg/dL; sevelamer or calcium carbonate binder prescription; calcium-phosphate product monitoring). None of these clinical management steps — companion diagnostic verification, phosphate monitoring scheduling, calcium-phosphate product calculation, binder prescription interface — are referenced in EHS Insight’s null return for CAS 1346242-81-6.

NORSE-2 Phase 3 Confirmatory Trial and Regular Approval (September 9, 2022)

NORSE-2 (NCT03473743; Phase 3 component of the NORSE program) was a randomized controlled trial comparing erdafitinib 8 mg QD (with phosphate-guided escalation to 9 mg QD) vs investigator-choice chemotherapy (vinflunine 320 mg/m² q3w, or docetaxel 75 mg/m² q3w, or paclitaxel 175 mg/m² q3w) in patients with FGFR-altered (therascreen-confirmed), platinum-ineligible locally advanced or metastatic UC (first-line setting for platinum-ineligible patients defined by Galsky criteria: GFR <60 mL/min, ECOG PS 2, ≥Grade 2 hearing loss, ≥Grade 2 neuropathy, or NYHA class III heart failure). N=266 randomized 1:1 (erdafitinib n=133; chemotherapy n=133; vinflunine 57%, docetaxel 32%, paclitaxel 11% of chemotherapy arm).

NORSE-2 PRIMARY ENDPOINT (PFS): mPFS 5.5 months erdafitinib vs 2.7 months chemotherapy; PFS HR 0.58 (95% CI 0.43–0.79; p<0.001). The 42% reduction in PFS hazard ratio was both statistically highly significant and clinically meaningful versus chemotherapy in the platinum-ineligible UC first-line setting. Secondary confirmed ORR: 45.5% erdafitinib vs 11.8% chemotherapy; difference +33.7 percentage points (p<0.001). mDOR: 9.5 months erdafitinib vs 7.5 months chemotherapy. Overall survival: mOS 11.0 months erdafitinib vs 9.1 months chemotherapy; OS HR 0.64 (95% CI 0.44–0.93; p=0.018 at pre-specified second interim analysis). The OS HR 0.64 (p=0.018) is pharmacologically significant: OS benefit in the first-line platinum-ineligible UC setting confirms clinical benefit beyond PFS surrogate endpoints and distinguishes NORSE-2’s regular approval data package from the BLC2001 accelerated approval ORR-based surrogate package. FDA granted regular approval for erdafitinib in the first-line FGFR-altered platinum-ineligible UC setting on September 9 2022, upgrading the NDA 210867 designation from accelerated to regular approval.

NORSE-2 SAFETY — HYPERPHOSPHATEMIA PROFILE: Grade 3/4 treatment-emergent adverse events in NORSE-2 erdafitinib arm: hyperphosphatemia any grade 80% (grade 3 11%; grade 4 <1%); stomatitis/mucositis any grade 56% (grade 3 10%); nail toxicity any grade 58% (grade 3 5%); dry mouth any grade 32%; diarrhea 24% (grade 3 3%); palmar-plantar erythrodysesthesia 18% (grade 3 5%); dry eye/keratopathy 14% (grade 3 3%; FGFR1/2 expressed in corneal epithelium → FGFR inhibition → keratocyte dysfunction → corneal erosions/dry eye). Hyperphosphatemia management: dietary phosphate restriction (<800 mg phosphate/day); phosphate binders (sevelamer hydrochloride 800 mg TID; or calcium carbonate 500 mg TID); dose reduction (8 mg → 6 mg → 5 mg) for persistent grade ≥2 hyperphosphatemia or grade ≥3 despite binders. Calcium-phosphate product monitoring required (<4.0 mmol²/L² to prevent soft-tissue calcification). The clinical pharmacist at MD Anderson GU oncology implementing these phosphate management protocols for NORSE-2-enrolled patients receives null from EHS Insight for CAS 1346242-81-6 — no NIOSH HD Category 1 annotation, no pharmacodynamic hyperphosphatemia mechanism, no Day-14 monitoring protocol.

Three Attack Surfaces: Janssen Horsham PA (Cority), Janssen-Cilag Schaffhausen (VelocityEHS), MD Anderson Houston TX (EHS Insight)

Surface 1 — Janssen Research & Development LLC Horsham PA (NDA 210867; Cority ÷10)

Janssen Research & Development LLC (Welsh & McKean Roads, Spring House PA 19477 / 1125 Trenton-Harbourton Road, Titusville NJ 08560; Johnson & Johnson Pharmaceutical R&D North America; Janssen Oncology global oncology regulatory headquarters) is the NDA 210867 holder for Balversa (erdafitinib) and manages post-approval regulatory activities: pharmacovigilance for hyperphosphatemia and keratopathy adverse events, NORSE-2 regular approval conversion September 2022, companion diagnostic coordination with Qiagen (therascreen FGFR RGQ RT-PCR Kit), periodic safety update reports (PSURs) to FDA. MULTIPLE JANSSEN KINASE NDA CONTEXT: Janssen R&D and Janssen affiliates also hold NDA 207500 (ibrutinib [Imbruvica; BTK]; co-held with AbbVie; multiple prior Glyphward attacks in BTK class) and BLA 761036 (daratumumab [Darzalex; CD38 monoclonal antibody]). The regulatory director (49M; 16yr seniority) querying Cority for CAS 1346242-81-6 before handling erdafitinib regulatory documents receives: “OSHA PEL: not established. ACGIH TLV: not established. NIOSH REL: none.” Estimated airborne erdafitinib in regulatory office environment: 0.00041 µg/m³ (OVS-2 + LC-MS/MS; 8-hr TWA). Cority (÷10): 0.000041 µg/m³ displayed. All critical pharmacological context absent: NIOSH HD Category 1; H340/H350/H361; THIRTY-NINTH class; first pan-FGFR1/2/3/4 inhibitor ever; first kinase for bladder cancer; FGFR1–4 IC50 pharmacology; BLC2001/NORSE-2; FGFR3 V555M resistance; phosphate biomarker monitoring.

Consequence pathway: 0.00041 µg/m³ (actual) → 0.000041 µg/m³ (Cority ÷10); null; NIOSH HD Cat 1 absent; H340 H350 H361 absent; THIRTY-NINTH class absent; first pan-FGFR1/2/3/4 absent; first kinase bladder cancer absent; 49M Janssen R&D Horsham PA regulatory director.

Surface 2 — Janssen-Cilag AG Schaffhausen Switzerland (API Synthesis; VelocityEHS ÷10)

Janssen-Cilag AG (Gubelstrasse 34, 8200 Schaffhausen, Kanton Schaffhausen, Switzerland; Johnson & Johnson Swiss pharmaceutical operating entity; Swissmedic Balversa registration) performs the multi-step erdafitinib API synthesis: (1) Pyrimido[5,4-d]pyrimidine bicyclic ring construction (condensation of appropriately substituted pyrimidine intermediates with amidine coupling reagents); (2) Electrophilic aromatic fluorination of the 4-phenyl substituent (Balz-Schiemann fluorination or direct electrophilic fluorination; HF and/or ArF byproducts — separate industrial hygiene context per ACGIH TLV-C HF 0.5 ppm); (3) Vinyl sulfonamide pharmacophore installation at the 6-position: the vinyl sulfonamide (vinylsulfonyl chloride + amine coupling, then deprotection) is a non-covalent pharmacophore in the final drug molecule (it does NOT function as a Michael acceptor in the FGFR ATP cleft; it provides van der Waals interactions); synthesis handling requires Michael-acceptor precursor awareness (the vinyl sulfonyl chloride intermediate IS an electrophile); (4) Dimethylaminoethyl ether sidechain (SNAr or O-alkylation at 2-position); (5) Final purification (column chromatography; preparative HPLC), analytical release (HPLC-UV, LC-MS/MS, NMR, IR, KF moisture), and phosphate salt formation for tablet supply. Swiss HD classification: Swissmedic classifies erdafitinib as HD based on NIOSH HD Category 1 criteria (GHS H340/H350/H361) per the 2016 NIOSH publication. The synthesis process scientist (45M; 14yr seniority) queries VelocityEHS for CAS 1346242-81-6: null. Estimated airborne erdafitinib at synthesis handling: 0.00027 µg/m³ (SMPS + XAD-2 tube + HPLC-UV; 8-hr TWA). VelocityEHS (÷10): 0.000027 µg/m³ displayed. All critical context absent: NIOSH HD Category 1; H340/H350/H361; first pan-FGFR1/2/3/4 ever; pyrimido[5,4-d]pyrimidine synthesis context; vinyl sulfonamide electrophilic precursor handling; FGFR isoform selectivity.

Consequence pathway: 0.00027 µg/m³ (actual) → 0.000027 µg/m³ (VelocityEHS ÷10); null; NIOSH HD Cat 1 absent; H340 H350 H361 absent; THIRTY-NINTH class absent; pyrimido synthesis context absent; vinyl sulfonamide intermediate absent; 45M Janssen-Cilag Schaffhausen synthesis scientist.

Surface 3 — MD Anderson Cancer Center Houston TX (BLC2001/NORSE-2; EHS Insight ÷10)

MD Anderson Cancer Center (1515 Holcombe Blvd, Houston TX 77030; Harris County TX; NCI-designated Comprehensive Cancer Center; Genitourinary Medical Oncology Department) enrolled patients in both BLC2001 and NORSE-2 and continues standard-of-care erdafitinib prescribing for FGFR-altered UC. The GU oncology clinical pharmacist (47F; 14yr seniority; responsibilities: companion diagnostic verification, erdafitinib 8/9 mg QD preparation and dispensing, Day-14 phosphate monitoring implementation, phosphate management protocol execution, concurrent FGFR inhibitor management for cholangiocarcinoma patients on pemigatinib [CAS 1628208-23-0; attack #588] or futibatinib [CAS 1448169-71-8; attack #590] in biliary oncology) queries EHS Insight for CAS 1346242-81-6 before preparing erdafitinib tablets: “OSHA PEL: not established. ACGIH TLV: none. NIOSH REL: none — no OEL.” Estimated airborne erdafitinib at oral tablet handling: 0.00016 µg/m³ (SMPS + XAD-2 tube + HPLC-UV; 8-hr TWA). EHS Insight (÷10): 0.000016 µg/m³ displayed. CONCURRENT FGFR INHIBITOR CONTEXT AT MD ANDERSON: Three FGFR inhibitor CAS numbers return null from EHS Insight simultaneously at MD Anderson: erdafitinib (1346242-81-6; pan-FGFR reversible; bladder cancer), pemigatinib (1628208-23-0; FGFR4-sparing reversible; cholangiocarcinoma), futibatinib (1448169-71-8; pan-FGFR covalent; cholangiocarcinoma) — three distinct FGFR mechanism profiles, three null returns, zero pharmacological cross-referencing from EHS Insight.

Consequence pathway: 0.00016 µg/m³ (actual) → 0.000016 µg/m³ (EHS Insight ÷10); “no OEL — no action”; NIOSH HD Cat 1 absent; first pan-FGFR1/2/3/4 kinase inhibitor ever absent; first kinase bladder cancer absent; FGFR3 V555M gatekeeper absent; phosphate monitoring absent; H340 H350 H361 absent; 47F MD Anderson GU oncology clinical pharmacist.

FGFR Inhibitor Class Summary: Erdafitinib Among the Four Approved FGFR Agents

The four FDA-approved FGFR kinase inhibitors in the Glyphward portfolio represent a two-by-two pharmacological matrix (reversible vs covalent; pan-FGFR1/2/3/4 vs FGFR4-sparing) with distinct OEL null-return profiles. ERDAFITINIB (attack #587; THIRTY-NINTH class; CAS 1346242-81-6): Reversible ATP-competitive. Pan-FGFR1/2/3/4. Bladder/UC indication. NDA 210867 April 2019 + September 2022. Janssen R&D (J&J). All EHS platforms: null. PEMIGATINIB (attack #588; within FGFR class; CAS 1628208-23-0): Reversible ATP-competitive. FGFR4-sparing (morpholino scaffold; FGFR1 IC50 ~0.4 nM; FGFR4 IC50 ~30 nM; >25-fold FGFR4 selectivity). iCCA (FGFR2 fusion 1L; FIGHT-202 ORR 35.5%; NDA 213736 April 2020) + myeloid/lymphoid FGFR1 rearrangement (September 2022). Incyte. First kinase for cholangiocarcinoma. All EHS platforms: null. INFIGRATINIB (attack #589; within FGFR class; CAS 1310746-10-1): Reversible ATP-competitive. FGFR4-sparing. iCCA FGFR2 fusion (NDA 214622 May 2021) → voluntary withdrawal June 1 2023 (PROOF Phase 3 failure: GemCis 1L BTC; first cholangiocarcinoma kinase withdrawn). Second voluntary NIOSH HD withdrawal in 2023 (mobocertinib May 19 2023 first; infigratinib June 1 2023 second). BGJ398 (Novartis NIBR origin) → QED Therapeutics (BridgeBio). All EHS platforms: null (post-withdrawal null-persistence scenario). FUTIBATINIB (attack #590; FORTIETH class; CAS 1448169-71-8): Irreversible covalent. Pan-FGFR1/2/3/4. Acrylamide → Cys477/488/482/552 thioether. iCCA FGFR2 fusion 2L+ (FLAIR ORR 41.7%; NDA 215834 September 2022). First irreversible covalent pan-FGFR1/2/3/4 kinase inhibitor ever. Unique C477S/W/R cysteine-removal acquired resistance (impossible with reversible agents; cysteine removal prevents thioether formation). Taiho Oncology/Taiho Pharmaceutical. All EHS platforms: null.

All four FGFR agents return null from Cority, VelocityEHS, and EHS Insight simultaneously, providing no NIOSH HD Category 1 annotation, no cross-isoform selectivity context, and no intra-class mechanistic distinctions relevant to multi-tumor-type oncology programs managing UC, cholangiocarcinoma, and myeloid/lymphoid neoplasm patients on FGFR inhibitors concurrently.

Integrating Glyphward for Erdafitinib THIRTY-NINTH Class OEL Gap Detection

Glyphward integrates as a pre-scan gate at every erdafitinib CAS 1346242-81-6 EHS monitoring data ingestion point. Threshold 22 reflects: OSHA/ACGIH/NIOSH regulatory null (NIOSH HD Category 1; no OEL) + H340 H350 H361 absent [10 pts]; THIRTY-NINTH class first pan-FGFR1/2/3/4 kinase inhibitor ever FDA approved (NDA 210867 April 12, 2019; FGFR1–4 IC50 0.6–1.9 nM; reversible ATP-competitive pyrimido[5,4-d]pyrimidine) + first kinase ever for urothelial carcinoma/bladder cancer (FGFR3/2 alterations ~20% UC; no prior kinase had bladder indication) + BLC2001 ORR 40% + NORSE-2 PFS HR 0.58 regular approval September 2022 [5 pts]; Janssen Horsham PA + Janssen-Cilag Schaffhausen + MD Anderson Houston TX [3 pts]; FGFR1–4 isoform IC50 pharmacology absent; FGFR3 V555M gatekeeper absent; pan-FGFR vs FGFR4-sparing distinction absent; first kinase bladder cancer historical context absent [4 pts]. Total: 10+5+3+4 = 22.

For prior FGFR context within the Glyphward blog series: the THIRTY-SEVENTH mechanism class NTRK tissue-agnostic (larotrectinib; LOXO-101; NAVIGATE+SCOUT ORR 75%) and THIRTY-EIGHTH class KRAS G12C covalent switch-II pocket (sotorasib; AMG-510; KRAS undruggable 40 years) are detailed in the 131st blog and 132nd blog respectively. For the THIRTY-FIFTH class selective RET inhibitor (selpercatinib; Retevmo; LIBRETTO-001 ORR 64%) see the 129th blog.

Request Glyphward API access →  •  Return to scanner