Casgevy
exagamglogene autotemcel
Species
Mouse
Longest study
20 wk
Genotox
Not conducted Not conducted
Carcinogenicity
Not conducted Not conducted
Repro
Not conducted Not conducted
First dose
Not conducted Not stated publicly
Speciesopen answer page →
Species
Mouse only (irradiated NSG immunodeficient xenotransplant mice)
“The toxicity of exa-cel has been evaluated in an in vivo combined GLP toxicity and tumorigenicity study with a duration of 20 weeks in NSG mice.”
Why that species
No relevant disease model; NSG xenotransplant mice support engraftment of human edited cells (most relevant model)
“No in vivo proof-of-concept studies were conducted as no relevant animal disease model for TDT and SCD exists that could be used for evaluation of exa-cel.”
Surrogate molecule
No surrogate; clinical-representative edited human CD34+ cells from healthy donors (surrogate gRNA/cells deemed not relevant)
“The test material used in these studies are representative of test material used in the clinic and derived from healthy donors.”
Species: general tox
Irradiated immunodeficient NSG mouse only (xenotransplanted edited human CD34+ cells)
“An in vivo toxicology and tumorigenicity study of SPY101-RNP edited CD34+ hHSPCs from healthy donors was conducted in irradiated NSG mice.”
Species: DART
Not conducted (not warranted given product characteristics and safety profile)
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
Species: carcinogenicity
Not conducted; tumorigenicity assessed in irradiated NSG mice instead
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
Surrogate used in
No surrogate used; edited cells from healthy donors representative of clinical material
“The test material used in these studies are representative of test material used in the clinic and derived from healthy donors.”
Study lengthopen answer page →
Longest pivotal study
20 weeks
“Moreover, combined GLP toxicity and tumorigenicity study has been conducted in NSG mice with a duration of 20 weeks.”
Pivotal GLP studies
GLP 20-week single-dose IV toxicity/tumorigenicity study and GLP biodistribution study (8 and 20 weeks) in irradiated NSG mice
“Both GLP studies were conducted as single-dose studies using the intravenous (IV) route reproducing the clinical use of exa-cel.”
Route in tox
Intravenous (single injection after total body irradiation)
“Route of Administration IV injection (0.25 mL/mouse) at Day 0”
Skipped studiesopen answer page →
Genotoxicity
No conventional studies; specialized on-/off-target editing (GUIDE-seq, in silico, hybrid capture) and chromosomal aberration assessments
“No evidence of off-target editing was observed for in-silico and GUIDE-seq identified candidate off-target sites which were assessed by hybrid capture sequencing analyses.”
Carcinogenicity
Not conducted (tumorigenicity evaluated in 20-week NSG study; no tumors)
“No carcinogenicity studies were conducted with CTX001.”
Reproductive & developmental
Not conducted (no reproductive organ findings, low gonadal distribution, non-integrating editing)
“No DART studies were conducted with CTX001.”
Safety pharmacology
No separate studies; selected parameters assessed in GLP toxicity study with no effects
“However, selected safety pharmacology parameters, were investigated in the GLP toxicity study and no effects on CNS, cardiac, respiratory, renal or gastrointestinal systems were observed.”
Genotox: Ames
Not conducted (conventional mutagenicity studies not done)
“Conventional mutagenicity and carcinogenicity studies have not been conducted for CASGEVY.”
Genotox: in vitro mammalian
Specialized in vitro assays in edited CD34+ cells: no off-target editing (GUIDE-seq, hybrid capture) and no chromosomal aberrations (karyotyping, LR-PCR)
“No evidence of off-target editing was observed for in-silico and GUIDE-seq identified candidate off-target sites which were assessed by hybrid capture sequencing analyses. No chromosomal aberrations for SPY101-RNP edited CD34+ hHSPCs from healthy donors were observed based on karyotyping, long-range PCR sequencing, and hybrid capture sequencing analyses.”
Genotox: in vivo
Not conducted (conventional mutagenicity studies not done)
“Conventional mutagenicity and carcinogenicity studies have not been conducted for CASGEVY.”
Carcinogenicity model
Tumorigenicity in irradiated NSG mice, single IV dose followed 20 weeks: no tumor formation
“Mice received single IV administration of 1 x 106 cells/mouse and were followed for 20 weeks. There were no significant adverse findings or tumor formation.”
Carcinogenicity timing
Not conducted: not warranted (product characteristics and safety profile)
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
hERG / QT
No safety pharmacology studies conducted
“There were no safety pharmacology studies conducted with CASGEVY.”
First doseopen answer page →
First-in-human starting dose
Not stated in public reviews
Why it's blank
The public FDA and EMA reviews of Casgevy don't say how the first human dose was chosen. That rationale is in the IND, which is confidential, so it is missing for most approved products.
How it's usually set for gene therapies
FDA expects preclinical proof-of-concept and toxicology studies to identify active dose levels and support the choice of starting dose and escalation. The 2013 guidance gives no numeric formula or scaling rule for the starting dose.
“Selection of potential starting dose level, dose-escalation schedule, and dosing regimen for clinical trials.”
✓ VerifiedFDA 2013: Preclinical Assessment of Investigational Cellular and Gene Therapy…, p. 7Source ↗
This product's starting point
NOAEL / key findings: 1x10^6 cells/mouse well tolerated; minor non-adverse red cell changes and spleen/thymus round-cell infiltration; no tumors (EMA assessment report, p. 44)
NOAEL / key findings
1x10^6 cells/mouse well tolerated; minor non-adverse red cell changes and spleen/thymus round-cell infiltration; no tumors
“The single dose GLP toxicology study with an observation period of 20 weeks revealed that the single IV injection of 1 x 106 cells/animal of unedited or SPY101-RNP-edited CD34+ hHSPCs into non-lethally irradiated NSG mice was well-tolerated.”
Risksopen answer page →
Immune & cytokine safety
In vitro innate immune (type I IFN) assay showed no activation, but FDA deemed it uninterpretable (positive control failed)
“The study results were uninterpretable since the positive control group of R848 CD34+ cells did not show high levels of expression of IFIT1, TIG1, OAS1, and IL-6 transcripts, which were observed in R848 PBMC cohort.”
Biodistribution & persistence
Human DNA highest in bone marrow, then spleen, blood, lung, liver, kidney; minimal in gonads; stable ~87% editing at 20 weeks
“Human DNA was detected in most of the examined tissues in mice that received a single IV administration of 1 x 106 cells/mouse, with the highest levels detected in the bone marrow, followed by spleen, blood, lung, liver, and kidney.”
Modality-specific risk
Off-target editing, large on-target indels and chromosomal translocations; NSG tumorigenicity model validity questioned
“Thus, the potential genotoxic/tumorigenic risk of exa-cel is solely judged by the in vitro studies investigating off-target gene editing and chromosomal translocations.”
Anti-drug antibodies: animals
Not stated in public reviews
Not reported in the public reviews. Reviewers summarise selectively, so this usually means it wasn't discussed publicly, not that it wasn't studied.
Anti-drug antibodies: patients
Not stated in public reviews
Not reported in the public reviews. Reviewers summarise selectively, so this usually means it wasn't discussed publicly, not that it wasn't studied.
Label warnings
- Neutrophil Engraftment Failure
- Delayed Platelet Engraftment
- Hypersensitivity Reactions
- Off-Target Genome Editing Risk
CytopeniasThrombocytopeniaHypersensitivity & infusion reactions
US prescribing information, Warnings and PrecautionsDailyMed ↗
Tox packageopen answer page →
Before first-in-human
First-in-human: where
Not stated in public reviews
Not reported in the public reviews. Reviewers summarise selectively, so this usually means it wasn't discussed publicly, not that it wasn't studied.
Repeat-dose tox before first-in-human
Not stated in public reviews
Not reported in the public reviews. Reviewers summarise selectively, so this usually means it wasn't discussed publicly, not that it wasn't studied.
First-in-human starting dose
Not stated in public reviews
Why it's blank
The public FDA and EMA reviews of Casgevy don't say how the first human dose was chosen. That rationale is in the IND, which is confidential, so it is missing for most approved products.
How it's usually set for gene therapies
FDA expects preclinical proof-of-concept and toxicology studies to identify active dose levels and support the choice of starting dose and escalation. The 2013 guidance gives no numeric formula or scaling rule for the starting dose.
“Selection of potential starting dose level, dose-escalation schedule, and dosing regimen for clinical trials.”
✓ VerifiedFDA 2013: Preclinical Assessment of Investigational Cellular and Gene Therapy…, p. 7Source ↗
This product's starting point
NOAEL / key findings: 1x10^6 cells/mouse well tolerated; minor non-adverse red cell changes and spleen/thymus round-cell infiltration; no tumors (EMA assessment report, p. 44)
Longest repeat-dose systemic studies
Longest repeat-dose study and recovery
No repeat-dose study (single administration)
“A repeat-dose toxicity study was not performed and is not required as exa-cel is intended for single-dose administration only.”
Pivotal GLP studies
GLP 20-week single-dose IV toxicity/tumorigenicity study and GLP biodistribution study (8 and 20 weeks) in irradiated NSG mice
“Both GLP studies were conducted as single-dose studies using the intravenous (IV) route reproducing the clinical use of exa-cel.”
DART package
Species: DART
Not conducted (not warranted given product characteristics and safety profile)
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
Reproductive & developmental
Not conducted (no reproductive organ findings, low gonadal distribution, non-integrating editing)
“No DART studies were conducted with CTX001.”
PPND / ePPND: when
Not conducted: DART not warranted (product characteristics and safety profile)
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
Carcinogenicity
Species: carcinogenicity
Not conducted; tumorigenicity assessed in irradiated NSG mice instead
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
Carcinogenicity model
Tumorigenicity in irradiated NSG mice, single IV dose followed 20 weeks: no tumor formation
“Mice received single IV administration of 1 x 106 cells/mouse and were followed for 20 weeks. There were no significant adverse findings or tumor formation.”
Carcinogenicity timing
Not conducted: not warranted (product characteristics and safety profile)
“Developmental and reproductive toxicity studies and carcinogenicity studies were not conducted with CASGEVY. These studies are not warranted based on the characteristics and safety profile of the product.”
Safety marginopen answer page →
Safety margin at NOAEL
NSG mouse toxicity dose gave only a 1.6-fold margin over the maximum human cell dose
“The dose level of 1 x 106 CD34+ hHSPCs/mouse administered in the single dose toxicity study is acceptable, despite the fact that there is only a 1.6-fold safety margin as compared to the max. human dose level of 20 x 106 cells/kg.”
NOAEL / key findings
1x10^6 cells/mouse well tolerated; minor non-adverse red cell changes and spleen/thymus round-cell infiltration; no tumors
“The single dose GLP toxicology study with an observation period of 20 weeks revealed that the single IV injection of 1 x 106 cells/animal of unedited or SPY101-RNP-edited CD34+ hHSPCs into non-lethally irradiated NSG mice was well-tolerated.”
Efficacy evidenceopen answer page →
Efficacy evidence
Edited human CD34+ cells engrafted in NSG mice retained >90% on-target editing at 16 weeks
“At 16 weeks post-administration, engrafted cells in NSG mice demonstrated >90% on-target editing frequencies in whole blood, bone marrow, and spleen.”
Why that model
No relevant animal disease model of TDT or SCD exists
“No in vivo proof-of-concept studies were conducted as no relevant animal disease model for TDT and SCD exists that could be used for evaluation of exa-cel.”
Reviewer's view of the efficacy data
EMA: pharmacodynamic studies adequate to show beneficial editing and HbF levels despite lack of a disease model
“Overall, the primary pharmacodynamic studies provided adequate evidence that potential beneficial levels of genome editing and HbF expression can be reached, despite the lack of a relevant disease animal model.”
Flexibilityopen answer page →
Flexibility accepted
Efficacy modelAcceptedBeyond the letter
No in vivo proof-of-concept study; no disease model for TDT or SCD, so engraftment and editing persistence studied in NSG xenotransplant mice
Why: No relevant animal disease model; in vitro data showed beneficial editing and HbF expression levels · No relevant species
“No in vivo proof-of-concept studies were conducted as no relevant animal disease model for TDT and SCD exists that could be used for evaluation of exa-cel.”
“Overall, the primary pharmacodynamic studies provided adequate evidence that potential beneficial levels of genome editing and HbF expression can be reached, despite the lack of a relevant disease animal model.”
CarcinogenicityAcceptedBeyond the letter
No carcinogenicity studies; NSG toxicity/tumorigenicity study of limited validity, so tumorigenic risk addressed by in silico and in vitro editing analyses
Why: Off-target editing and chromosomal aberration risk sufficiently addressed using in silico and in vitro approaches · Weight of evidence
“However, the significance of the in vivo model used for addressing the tumorigenic risk associated with potential off-target editing and/or chromosomal translocations induced during the editing process of the CD34+ hHSPCs is limited, as the validity of the model remains questionable.”
“Thus, the tumorigenic risk of exa-cel needs to be addressed primarily by the evaluation of the potential of SPY101-RNP to induce off-target editing and chromosomal aberrations, which has been sufficiently addressed using in silico and in vitro approaches.”
GenotoxicityAcceptedBeyond the letter
No dedicated chromosomal translocation studies (e.g. CAST-Seq); translocation risk estimated indirectly from large-insertion rates
Why: Other methods such as CAST-Seq unlikely to detect translocations · Weight of evidence
“However, as the likelihood of detecting chromosomal translocations using other methods such as CAST-Seq is expected to be low, omission of additional chromosomal translocation studies is acceptable.”
Safety marginCriticised but acceptedBeyond the letter
Single-dose NSG toxicity study dose gave only a 1.6-fold margin over the maximum human cell dose
Why: Main safety concern is off-target editing and genotoxicity, addressed in the genotoxicity evaluation · Weight of evidence
“The dose level of 1 x 106 CD34+ hHSPCs/mouse administered in the single dose toxicity study is acceptable, despite the fact that there is only a 1.6-fold safety margin as compared to the max. human dose level of 20 x 106 cells/kg.”
“However, the main safety concern of exa-cel is associated with editing at critical potential off-target sites and associated genotoxicity that might induce tumorigenic events. This risk is addressed mainly in the non-clinical genotoxicity evaluation, which included evaluation of both on-target and off-target editing.”
DARTAcceptedBeyond the letter
No developmental and reproductive toxicity studies
Why: No reproductive organ findings, very low distribution to reproductive organs; acceptable on product characteristics and safety profile · Weight of evidence · Low-risk modality
“No DART studies were conducted with CTX001. The applicant’s rationale for not conducting DART studies include: i) no CTX001-related pathologic findings were observed in reproductive organs in the completed toxicology study (Study Report No. 1016-2465), ii) very low distribution of CTX001 to reproductive organs was reported in the completed biodistribution study (Study Report No. 1016-2475),”
“Reviewer comment: This is acceptable based on the product characteristics and safety profile.”
Similar precedents
LyfgeniaGene therapy · lovotibeglogene autotemcel · HBB T87Q (LVV, ex vivo HSC)Rodent only · 26 wkZyntegloGene therapy · betibeglogene autotemcel · HBB T87Q (LVV, ex vivo HSC)Rodent only · 26 wkFayuviGene therapy · rebisufligene etisparvovec · SGSH (AAV9)Rodent only · 26 wkLenmeldyGene therapy · atidarsagene autotemcel · ARSA (LVV, ex vivo HSC)Rodent only · 52 wkStrimvelisGene therapy · autologous CD34+ cells transduced with ADA (no INN) · ADA (gamma-retroviral, ex vivo HSC)Rodent only · 16 wkWaskyraGene therapy · etuvetidigene autotemcel · WAS (LVV, ex vivo HSC)Rodent only · 64 wk