FDA Approval Summary: Ciltacabtagene Autoleucel for Relapsed or Refractory Multiple Myeloma

Abstract

In February 2022, the U.S. Food and Drug Administration approved ciltacabtagene autoleucel, a chimeric antigen receptor (CAR) T cell therapy targeting the B-cell maturation antigen (BCMA), for adult patients with relapsed/refractory multiple myeloma (RRMM) after ≥4 lines of therapy including an immunomodulatory agent, a proteasome inhibitor, and an anti-CD38 monoclonal antibody. Approval was based on overall response rate (ORR), complete response (CR) rate and duration of response (DOR) in 97 adult patients in a single-arm, open-label, multicenter phase 2 trial (CARTITUDE-1 [NCT03548207]). Patients received a single infusion of ciltacabtagene autoleucel, preceded by lymphodepleting chemotherapy. Of the 97 patients evaluable, ORR was 97.9% (95% CI: 92.7, 99.7) with stringent CR rate of 78.4% (95% CI: 68.8, 86.1). After median follow-up of 18 months, median DOR was 21.8 months (95% CI: 21.8, not estimable [NE]) in responders (PR or better) and NE (95% CI: 21.8 months, NE) in patients who achieved stringent CR. Serious adverse reactions occurred in 55% of 97 patients evaluated for safety. Grade 3 or higher cytokine release syndrome and neurologic toxicities occurred in 5% and 11%, respectively, leading to a Risk Evaluation and Mitigation Strategy. Neurologic toxicities included immune effector cell-associated neurologic syndrome (ICANS) typically seen with CAR-T products, parkinsonism, peripheral neuropathy, cranial nerve palsies and Guillain-Barre syndrome (GBS). One fatal case of hemophagocytic lymphohistiocytosis/macrophage activation syndrome occurred. Prolonged and recurrent grade 3 or 4 cytopenias occurred; a single patient required hematopoietic stem cell rescue.

Introduction

Multiple myeloma is the second most common hematologic malignancy in the US, accounting for 1.8% of all cancers and 17% of all hematologic malignancies¹. Patients who are refractory to major classes of available anti-myeloma therapies such as triple class refractory (refractory to a proteasome inhibitor [PI], an immunomodulatory agent [IMiD] and an anti-CD38 monoclonal antibody), or penta-refractory (refractory to 2 PIs, 2 IMiD agents, and an anti-CD38 monoclonal antibody) demonstrate low response rates to salvage therapies and have poor overall prognosis². Therefore, there is a substantial need for new therapies for treatment of patients with myeloma that is refractory to main classes of agents such as proteasome inhibitors, immunomodulatory agents and anti-CD38 antibodies.

Ciltacabtagene autoleucel is the second CAR T cell product approved for the treatment of multiple myeloma. On February 28, 2022, after a priority review, the Food and Drug Administration (FDA) granted regular approval to ciltacabtagene autoleucel for the treatment of adult patients with relapsed or refractory multiple myeloma after four or more prior lines of therapy, including an immunomodulatory agent, a proteasome inhibitor and an anti-CD38 monoclonal antibody. The efficacy of ciltacabtagene autoleucel was established in CARTITUDE-1 (NCT03548207), a single-arm, multicenter trial, based on the overall response rate (ORR), complete response (CR) rate, and the duration of response (DoR) in a heavily pre-treated patient population. Preliminary data from this trial was the basis for Breakthrough Therapy Designation (BTD) in multiple myeloma, which was granted in December of 2019. Herein, we provide a summary of the FDA clinical review and rationale for regular approval of this marketing application.

Drug Product

Ciltacabtagene autoleucel is a genetically modified T cell product consisting of autologous T cells transduced with a lentiviral vector (LVV) expressing a chimeric antigen receptor (CAR) targeting the B-cell maturation antigen (BCMA)³. The CAR is comprised of a human CD8 alpha signal peptide (CD8α SP), BCMA targeting domains (VHH1 and VHH2), human CD8 alpha hinge and transmembrane domain (CD8α hinge +TM), human CD137 cytoplasmic domain (4-1BB), and a human CD3 zeta cytoplasmic domain (CD3ζ). Binding of the anti-BCMA CAR to BCMA-expressing target cells leads to signaling through the CD3ζ and 4-1BB domains, and subsequent CAR T cell activation. Antigen-specific activation results in CAR T cell proliferation, cytokine secretion, and subsequent cytolytic killing of BCMA-expressing cells.

The recommended dose is a single infusion of 0.5 to 1.0 × 10⁶ CAR-positive T cells per kg of body weight. To facilitate CAR T cell engraftment and expansion, lymphodepleting chemotherapy, cyclophosphamide 300 mg/m² intravenously (IV) and fludarabine 30 mg/m² IV was administered for 3 days prior to infusion. Ciltacabtagene autoleucel was administered 2-4 days after completion of lymphodepleting chemotherapy. Lymphodepleting chemotherapy was to be delayed if a patient had significant toxicity from preceding bridging therapy. Repeat lymphodepletion was recommended if ciltacabtagene autoleucel had been delayed by more than 14 days following the first lymphodepleting chemotherapy course.

Clinical Trial Design

The basis of approval was a single-arm, open-label, multicenter phase 2 trial (CARTITUDE-1 [NCT03548207]) in adults with relapsed and refractory multiple myeloma, with a primary efficacy endpoint of ORR per independent review committee (IRC) based on International Myeloma Working Group (IMWG) Consensus Criteria, 2016⁴. Eligible patients had received at least three prior lines of therapy including an immunomodulatory agent, a proteasome inhibitor and anti-CD38 antibody with documented disease progression on or within 12 months of the last line of therapy. The study required an Eastern Cooperative Oncology Group performance status (ECOG PS) < 2, creatinine clearance (Cr Cl) ≥40mL/min/1.73 m² and hepatic transaminases ≤ 2.5 times the upper limit of normal. Study participants were required to have absolute neutrophil count (ANC) ≥750 cells/mm³ and platelet count of ≥50,000/mm³ in the absence of growth factor or transfusion support, respectively. All patients had measurable disease by IMWG 2016 criteria⁴ at enrollment.

Antimyeloma bridging therapy was permitted between leukapheresis and prior to initiation of lymphodepleting chemotherapy. Patients could only receive bridging with antimyeloma therapies that they had previously received. Following the 1^st six patients, who were required to be hospitalized for a minimum of 14 days following CAR-T cell infusion, the requirement for mandatory hospitalization was changed to 10 days after CAR T infusion to monitor for development of cytokine release syndrome (CRS) and neurotoxicity.

Results

Pharmacokinetics

Following infusion, ciltacabtagene autoleucel exhibited an initial expansion phase followed by a rapid decline and then a persistent phase with a slower decline subsequently. This pattern was observed in peripheral blood and bone marrow. Persistence was observed for all patients although high inter-individual pharmacokinetic variability was observed. For 65 patients in the trial with data available at the time of analysis, the median time for CAR transgene levels in the peripheral blood to return to pre-dose baseline was ~100 days (range 28 to 365 days) post-infusion.

Efficacy

Of the 113 patients who underwent leukapheresis, 97 patients treated with ciltacabtagene autoleucel at a dose range of 0.5-1 x10⁶ CAR+ T cells per kg body weight constitute the efficacy evaluable population. The median time from leukapheresis to product availability was 32 days (range: 27 to 66 days). Sixteen patients were not treated following leukapheresis, primarily due to progressive disease, death, and study withdrawal.

Seventy-three of 97 subjects received bridging therapy. Among the 73 patients, 36 patients experienced an increase in tumor burden following bridging therapy while 33 patients had a decrease in tumor burden. Fifteen of the 33 patients experienced a decrease of over 50%. Despite the decrease in tumor burden in some patients, no patient achieved a CR or better while on bridging therapy.

Table 1 summarizes the characteristics of the efficacy population. In the efficacy evaluable population (n=97), the median age was 61 years, 36% were ≥65 years. Eighteen percent of the efficacy evaluable population was Black or African American. Ninety-nine percent of the patients were refractory to an anti-CD38 antibody, 87.6% were triple class refractory, and 42.3% were penta-refractory. Although the trial eligibility allowed enrollment of patients after three or more prior lines of therapy, only 18% of the efficacy population had received three prior lines of therapy with remaining 82% having received ≥4 prior lines of therapy (Table 1).⁵

Table 1.

Characteristics of the efficacy population

Parameter	N=97
Time since initial diagnosis (years)
Median	5.9
Min, max	1.6, 18.2
International Staging System (ISS) stage at study entry n (%)
Stage I	61 (62.9)
Stage II	22 (22.7)
Stage III	14 (14.4)
Light chain type, n(%) at baseline(any)
Kappa light chain	15 (15.5)
Lambda light chain	9 (9.3%)
Biclonal	4 (4.1)
Immunoglobulin, n (%) At Baseline (Any)
Ig A	8 (8.2)
Ig G	57 (58.8)
Ig M	2 (2.1)
Ig D	2 (2.1)
Ig E	0
Not detected	0
Baseline cytogenetics risk n(%)
High risk	23 (23.7%)
Non-high risk	68 (70.1%)
Missing	6 (6.2%)
Presence of extramedullary plasmacytoma n (%)
Yes	13 (13.4%)
No	84 (86.6%)
Lytic bone disease n (%)
Yes	69 (71.1)
No	28 (28.9)
Tumor BCMA expression n (%)
<50%	5 (5)
≥50%	57 (59)
Unknown	35 (36.1%)
Number of prior antimyeloma regimens
Median (min, max)	6 (3, 18)
Distribution of prior antimyeloma regimens n (%)
3	17 (17.5)
4	16 (16.5)
5	15 (15.5)
>5	49 (50.5)
Prior stem cell transplant n (%)
Yes	87 (89.7)
1 prior transplant	70 (72.2)
>1 prior transplant	17 (17.5)
No	10 (10.3)
Prior refractory status n (%)
Immunomodulatory Agent (IMiD)	97 (100)
Proteasome inhibitor (PI)	97 (100)
Anti-CD38 antibodies	97 (100)
Daratumumab	94 (96.9)
Double refractory (IMiD and PI)	97 (100)
Triple refractory (IMiD, PI and anti-CD38)	97 (100)
Penta-refractory	81 (83.5)
Prior refractory to last regimen, n (%)	97 (100)

Data from Source: BLA clinical review memorandum 125746/0⁵

The ORR in the efficacy evaluable population was 97.9% (95% CI: 92.7%, 99.7%) (Table 2).⁶ The lower limit of the 95% confidence interval was greater than the pre-specified null hypothesis rate of 30% for ORR. The stringent complete response rate (sCR) was 78.4% (95% CI: 68.8%, 86.1%). In a subpopulation analysis that received bridging therapy, the ORR aligned with the efficacy evaluable population.

Table 2:

Efficacy Outcomes of the efficacy population

Efficacy Outcomes: N=97
Response per IRC according to IMWG 2016
Overall Response Rate, n (%)	95 (97.9%)
(sCR+VGPR+PR)
(95% CI)	(92.7, 99.7)
Stringent Complete response, n (%)	76 (78.4%)
(95% CI)	(68.8,86.1)
Very Good Partial Response, n (%)	16 (16.5%)
(95% CI)	(9.7,25.4)
Partial Response, n (%)	3 (3.1%)
(95% CI)	(0.6,8.8)
Estimated median DOR (95% CI), (months)
All responders	21.8 (21.8, NE)

Data from source: Package insert for ciltacabtagene autoleucel⁶

Overall, the ORR in the leukapheresis population (n=113) was 84.1% (95% CI: 76%, 90.3%) with a stringent CR rate of 67.3% (95% CI: 57.8%, 75.8%). The median duration of follow up for all patients was 18 months. The response duration tended to be longer in patients with sCR as compared to very good partial response (VGPR) and PR. Of the 95 patients who achieved any response, median duration of response was 21.8 months (95% CI: 21.8, NE) and an estimated 72.9% maintained a response for at least 12 months. The group of sCR did not reach its median DOR at the time of clinical cut-off.

Minimal residual disease (MRD), a secondary endpoint was assessed in the bone marrow samples using the ClonoSEQ assay. However, there was a 20.6% (20/97) calibration failure rate, or no clone was identified. This raised concerns regarding the reliability of the MRD response assessments. Thus, further analyses including impact of MRD on DOR was not performed. Therefore, the MRD data was not considered robust and not included in the label.

Time to event endpoints, such as progression-free survival (PFS) and overall survival (OS), were not assessed for efficacy. In a single arm trial these results are challenging to interpret as they may be subject to bias.

Safety

Safety was evaluated in all 97 patients who received an infusion of ciltacabtagene autoleucel at a dose range of 0.5-1 x10⁶ CAR+T cells per kg body weight in CARTITUDE-1. Additionally, safety data from other ongoing trials across the ciltacabtagene autoleucel clinical development program were reviewed to identify safety signals of interest and incorporated in the label.

All patients experienced at least one adverse reaction (AR). The most common ARs are shown in Table 3.⁶ Serious ARs were reported in 55% (53 of 97) of patients. The fatal AR rate was 9%. Causes of death included CRS with hemophagocytic lymphohistiocytosis (HLH/MAS) (n=1), AML (n=2), septic shock (n=1), neurologic toxicity (NT; n = 3) and infection (n=3). In 2 patients, death was multifactorial: one patient with infection (sepsis) had immune effector-cell associated neurologic syndrome (ICANS), one patient with NT [ICANS, cerebrovascular accident (CVA)] had pulmonary embolism.

Table 3.

Most common adverse reactions in at least 10% of patients (N=97)

Body System Organ Class AE	All Grades (%)	Grades 3 - 5 (Max Grade) (%)
Blood and lymphatic system disorders
Coagulopathy*	22	2
Febrile neutropenia	10	9
Cardiac disorders
Tachycardia *	27	1
Gastrointestinal disorders
Diarrhea*	33	1
Nausea	31	1
Constipation	22	0
Vomiting	20	0
General disorders & administration site conditions
Pyrexia	96	5
Fatigue *	47	7
Chills	33	0
Edema*	23	0
Immune system disorders
Cytokine Release Syndrome*	95	5
Hypogammaglobulinemia *	93	2
Infections and infestations*
Infections: pathogen unspecified*	41	19
Upper respiratory tract infection*	28	3
Viral infections*	23	7
Pneumonia *	14	13
Sepsis*	10	7
Metabolism and nutrition disorders
Decreased appetite	29	1
Musculoskeletal and connective tissue disorders
Musculoskeletal pain*	48	2
Nervous system disorders
Encephalopathy*	30	6
Headache	27	0
Dizziness*	23	1
Motor dysfunction *	16	3
Psychiatric disorders
Insomnia*	13	0
Respiratory, thoracic and mediastinal disorders
Cough*	39	0
Dyspnea*	23	3
Nasal congestion	15	0
Hypoxia	12	4
Vascular disorders
Hypotension*	51	10
Hypertension	19	6
Hemorrhage*	16	4

^*Includes group terms. Refer to prescribing information for definitions.

Data from source: Prescribing information of ciltacabtagene autoleucel ⁶

CRS was graded using the 2019 ASTCT criteria⁷; organ toxicity secondary to CRS was graded using the CTCAE v 5.0 criteria. CRS occurred in 95% of the treated patients and Grade ≥3 events occurred in 5%. Median time to CRS onset was 7 days (range 1 to 12 days). CRS resolved in all but 1 patient with a median time to resolution of 4 days (range 1 to 40 days). One patient had fatal CRS with a duration of 97 days. Overall, 71% received tocilizumab and/or corticosteroids for CRS management. Seventy percent (68 of 97), 25.7% (25/97) and 17% (18/97) received tocilizumab, corticosteroids and anakinra respectively for CRS management. Thirty-two patients had organ toxicity associated with CRS with 22 of these patients having ≥ Grade 3 organ toxicity. Organ toxicities associated with CRS included: hepatic (24%; manifest as elevated liver enzymes), cardiac (18.5%; arrythmias mainly), pulmonary (6.5%) and renal (4.4%). The Applicant switched between grading systems during the trial (used 2014 Lee criteria in phase 1b and 2019 ASTCT criteria during phase 2).^7,8 Therefore, some patients in phase 1b of the trial had neurologic signs/symptoms listed under CRS symptoms which were subsequently categorized as NT from the product.

HLH/MAS, a hyperinflammatory syndrome⁹, was reported in a single patient in CARTITUDE-1. Widespread hemophagocytic lymphohistiocytosis was demonstrated in multiple organs on autopsy. HLH/MAS has also been reported in additional patients who have received ciltacabtagene autoleucel in other clinical trials in multiple myeloma.

Neurologic toxicity included not only immune effector cell associated neurologic syndrome (ICANS) which is a constellation of classic signs and symptoms attributed to CAR-T therapy but also other toxicities including Parkinsonism, GBS, cranial nerve palsies and peripheral neuropathy. The majority of patients (70%; 68/97) had one or more neurologic and/or psychiatric system organ class adverse events (AE) and 25 patients (26%) were deemed to have neurologic toxicity (NT). Grade ≥ 3 toxicity occurred in 11 patients (11%). The most common (≥ 5%) neurologic signs/symptoms by FDA-group terms in all 97 patients treated with ciltacabtagene autoleucel include encephalopathy (30%), headache (27%), dizziness (23%), motor dysfunction (16%) aphasia (8%), ataxia (8%), peripheral neuropathy (7%) and tremor (6%). All but 1 patient had at least one symptom/sign (event) of NT start within 8 weeks of ciltacabtagene autoleucel infusion. The median time to the onset of first event was 8 days (range 2-101 days). Neurologic toxicity resolved in 15 of 25 patients (60%) with a median time to resolution of 8 days (range 2-208 days). Median duration of NT in all patients including those with fatal events and NT ongoing at death or last known alive date was 62 days (range 2 to 926 days). Of the 10 patients in whom NT did not resolve, 3 patients died of NT—one patient with parkinsonism, one with ICANS, and 1 with ICANS and CVA. Three patients had NT ongoing at time of death while 4 patients had NT ongoing at the last known alive date. Like CRS, different grading systems for ICANS were used during different phases of the clinical trial—CTCAE during phase 1b and ASTCT consensus grading during phase 2b.⁷ Overall, 5 additional cases of NT were identified upon FDA review, neurologic events were added in 4 patients already deemed to have NT and 2 patients had NT grade increase upon FDA adjudication including a change from grade 4 to grade 5 NT.

Categories of Neurotoxicity

ICANS:

ICANS occurred in 22 of 97 patients in CARTITUDE-1 with a median time-to-onset of 8 days (range 1-28 days). Seventeen of 22 patients (77%) had resolution of ICANS with a median time to resolution of 6 days (range 2-143 days). Two patients died of ICANS. Corticosteroids and antiseizure medications were used to manage ICANS.

Parkinsonism:

Five of 97 (5%) patients in CARTITUDE-1 had a constellation of several signs/symptoms consistent with parkinsonism with a median time to onset of 43 days (range 15-108 days). These patients had an admixture of parkinsonism (e.g., tremor, bradykinesia, gait disturbances, psychomotor retardation etc.) and non-parkinsonian (e.g., encephalopathy) neurologic signs/symptoms and, were considered to have NT distinct from ICANS. An additional patient with similar NT in another trial of ciltacabtagene autoleucel was reported at the time of the BLA submission. All 6 patients with parkinsonism were males with a median age of 60 years (range 44-77 years); all had CRS and 4 of 6 patients had grade 1 ICANS. Most patients (5 of 6) had Grade 3 or higher toxicity. Neurologic toxicity with parkinsonism did not resolve in any of the 5 patients in CARTITUDE-1. One patient died of the toxicity while the remainder had toxicity ongoing at death from other causes (n=2) or at last known alive day. Attempted treatment in these 6 patients included systemic chemotherapy, systemic glucocorticoids, intrathecal chemotherapy with or without corticosteroids, dopaminergic agents, plasmapheresis, intravenous immunoglobulin, siltuximab and dasatinib. No patient had resolution of toxicity. Brain autopsy results were available in 2 patients and were notable for focal gliosis of anterior caudate lobe, medial thalamus, basal ganglia, deep gray matter, and white matter, perivascular and parenchymal CD3+ T-cell infiltration in multiple brain regions including the basal ganglia and preservation of pigmented neurons in the substantia nigra and locus ceruleus. Preservation of pigmented neurons in the substantia nigra is in contrast to Parkinson Disease which is characterized by loss of these neurons in the substantia nigra.

Guillain Barre Syndrome (GBS):

A single fatal case of GBS has been reported to date in another trial of ciltacabtagene autoleucel. The patient reported symptoms consistent with the Miller-Fisher variant of GBS, polyradiculoneuritis, encephalopathy, speech disturbances and died despite treatment with intravenous immunoglobulins.

Peripheral neuropathy:

Six patients in CARTITUDE-1 developed peripheral neuropathy that were either sensory, motor or mixed sensorimotor neuropathies with a median time to onset of 62 days (4-136 days). Two of 6 patients experienced grade 3 neuropathy, and 3 of 6 patients had non-resolution of their AE. Peripheral neuropathy attributed to product has also been observed in other ongoing trials of ciltacabtagene autoleucel. Cranial nerve palsies and GBS have been reported along with peripheral neuropathy in several patients. It is unclear whether certain instances of peripheral neuropathy or facial nerve palsy could represent GBS or its variants in some patients.

Cranial nerve palsies:

Three patients in CARTITUDE-1 had cranial nerve palsy with several additional patients with this AE reported in other clinical trials of ciltacabtagene autoleucel. The 7^th cranial nerve is most commonly involved; palsies of the 3^rd, 5^th, and 6^th cranial nerve have also been reported. Bilateral cranial nerve palsies and recurrence of palsy after resolution have been reported. The median time to onset is variable; majority of patients appear to respond to corticosteroids.

Prolonged cytopenia was defined as any Grade 3 or 4 cytopenia (neutropenia, thrombocytopenia, lymphopenia, or anemia) that was persistent for more than 30 days after receiving ciltacabtagene autoleucel. Forty-one, thirty and twelve percent of 97 patients experienced prolonged thrombocytopenia, neutropenia, and lymphopenia, respectively, with significant number requiring growth factor support and/or transfusions for neutropenia and thrombocytopenia. Infections occurred in 25 patients (26%; 25/97) with prolonged neutropenia and/or lymphopenia. Eighty-four patients (87%; 84/97) had one or more recurrence of grade 3 or 4 cytopenia following recovery from 1^st episode of cytopenia to ≤ Grade 2 with some recurrences occurring many months following ciltacabtagene autoleucel infusion One patient underwent autologous hematopoietic stem cell transplant for prolonged thrombocytopenia.

Infections occurred in 57 patients (59%; 57/97) with Grade 3 or 4 infections in 22 patients. Five patients died of an infection. Hypogammaglobulinemia reported either as an AE or a laboratory IgG level of < 500mg/dl following ciltacabtagene autoleucel infusion occurred in 91 patients (94%; 91/97). Thirty-eight percent of hypogammaglobulinemic patients received IVIG.

Hypersensitivity reactions occurred in 5 patients (5%; 5/97); all grade 1.

Management of toxicities related to ciltacabtagene autoleucel is described in the prescribing information.⁶

Regulatory Insights

This is the second FDA approval of a BCMA targeted CAR T therapy for multiple myeloma. Approval was supported by results of a single arm, open label, multi-center study demonstrating ORR of 97.9% (95% CI: 92.7, 99.7) and sCR rate of 78.4% (95% CI: 68.8, 86.1) with durability in a relapsed and refractory population.

Approval Pathway

In granting regular approval to ciltacabtagene autoleucel in relapsed or refractory myeloma, the review team considered the magnitude of benefit observed, specifically the high ORR of 97.9%, sCR rate of 78.4% and the median DOR of 21.8 months in all responders, to constitute clinical benefit in this refractory population. Despite the serious and severe toxicity associated with ciltacabtagene autoleucel and lymphodepleting chemotherapy, this magnitude of treatment effect supported a favorable benefit-risk profile of ciltacabtagene autoleucel in the indicated patient population.

Indication Statement

The Applicant’s proposed indication for this product was the treatment of adult patients with multiple myeloma who have received at least three prior therapies, including an immunomodulatory agent, a proteasome inhibitor, and an anti-CD38 antibody. Although patients with R/R myeloma who had received at least three prior lines of therapy were eligible for the study, 82% of the efficacy evaluable population had received four or more lines of therapy with a median of 6 prior lines of therapy. Given the limited data to establish the benefit risk in patients who received 3 prior lines, the indication was restricted to patients after at least 4 prior lines of therapy.

Safety

This is the 1^st CAR T product label wherein CRS grading by the 2019 ASTCT criteria is used. Differences in CRS grading systems used during the different phases of the trial required re-adjudication of CRS events in Phase 1 part of the trial using the 2019 ASTCT criteria. Since organ toxicity is not incorporated in the 2019 ASTCT criteria, management of CRS in the label included a modification of the ASTCT grading criteria to include organ toxicity to be helpful to the end user. Similar to CRS, change in grading scale used for ICANS during the trial but re-adjudication was difficult given lack of information on components of the ASTCT grading system in some patients. Other neurologic toxicity attributed to the drug product was graded using the CTCAE criteria. Hence, the CTCAE grading criteria were used for all NT attributed to drug product. However, since management of NT in Section 2 of the label was specific to ICANS, the 2019 ASTCT criteria was used in this section to align with what was done in majority of patients in the trial for this particular kind of NT. The term Parkinsonism was used to describe the NT with clinical features of Parkinson Disease as opposed to terminology proposed by the Applicant to facilitate easy recognition of this unique NT by providers. NT events for duration of the available follow-up were included since some of the NT events were delayed in onset. During the clinical study, the Applicant introduced reduction in tumor burden as a mitigation strategy for parkinsonism. However, there was limited data at the time of this approval to support a specific mitigation strategy and was therefore not included in the label.⁵

The prescribing information for ciltacabtagene autoleucel contains boxed warnings for cytokine release syndrome (CRS), neurologic toxicities (ICANS, parkinsonism and GBS), HLH/MAS, and prolonged and/or recurrent cytopenia. Other Warnings and Precautions in the prescribing information include peripheral neuropathy and cranial nerve palsies under section on neurologic toxicities, hypersensitivity reactions, serious infections, hypogammaglobulinemia, secondary malignancies, and, due to the potential for neurologic events, restrictions on driving and other activities.

Risk Evaluation and Mitigation Strategy (REMS)

To ensure safety of the study participants, investigators and the study staff underwent training in the management of CRS and neurotoxicity. The protocol required that tocilizumab be available for CRS management prior to CAR T infusion and included detailed management guidelines for CRS and neurotoxicity. FDA determined that a Risk Evaluation and Mitigation Strategy (REMS) was necessary to ensure the safe use of ciltacabtagene autoleucel and maintain a favorable benefit-risk profile. Given the safety signals of parkinsonism, GBS, peripheral neuropathy, cranial nerve palsies, HLH/MAS and prolonged and recurrent cytopenias necessitating rescue hematopoietic stem cell transplantation, the Applicant’s proposed REMS training material was modified to include education material for providers regarding these toxicities in addition to CRS and ICANS.

The prescribing information for ciltacabtagene autoleucel recommends that patients receiving this product should be monitored at least daily for 10 days following infusion at a certified health care facility. Furthermore, patients should remain within proximity of that health care facility for at least 4 weeks following infusion to ensure prompt evaluation and management of toxicities. The feasibility and logistics of product handling at these facilities was considered by the manufacturing review team.

The safety data from this trial has to be considered in the context of the relatively strict eligibility criteria for the study population. Eighteen percent of the treated population was African American limiting the safety and efficacy data in this subgroup.¹⁰ Only 3% of the patients had creatinine clearance <45 ml/minute and 4% had an ECOG PS 2. Few patients (8%) ≥75 years or older were enrolled to characterize the efficacy and safety in this age group. The post marketing observational study described below will collect additional safety and efficacy data from patients treated with ciltacabtagene autoleucel in the real-world setting.

Post-marketing Studies

There are two main potential safety concerns related to the viral vector with ciltacabtagene autoleucel. The first is the possibility of vector integration into the host genome resulting in either activation of cellular proto-oncogenes or disruption of a tumor suppressor gene resulting in the development of secondary malignancies. The second concern is the generation of replication-competent lentivirus during the manufacturing process for ciltacabtagene autoleucel which may result in uncontrolled proliferation of genetically modified T cells. These concerns warrant long-term safety follow up post-treatment.

Therefore, a post-marketing requirement (PMR) study was issued that requires 15 years of follow-up for patients treated with ciltacabtagene autoleucel to assess its long-term toxicities. This observational PMR will also capture the incidence and severity of CRS, neurologic toxicities including Parkinsonism, HLH/MAS, prolonged cytopenia including the number of patients who require rescue stem cell transplantation and the outcomes in terms of hematopoietic reconstitution and survival.

Bridging Therapy

In general, there are differences that exist across programs in allowance of bridging therapy based on the study population, disease characteristics and manufacturing time. Key aspects of the protocol that ensured that the efficacy assessment based on ORR was not confounded by the effects of bridging include restricting bridging to therapies that were received prior to study enrollment and requirement for restaging post-bridging and prior to CAR T administration. Seventy-five percent of the treated population required bridging therapy, reflecting the refractory nature of the treated population and the median time from leukapheresis to product availability of 32 days. All 73 patients who received bridging therapy had measurable disease prior to lymphodepletion. Baseline efficacy assessments occurred after bridging therapy and prior to the start of the conditioning regimen to ensure all patients had measurable disease prior to receiving ciltacabtagene autoleucel.

Manufacturing Failure Rate

The product that met specifications for the CARTITUDE −1 study was referred to as ciltacabtagene autoleucel and the product specifications that were introduced after the completed study and during the review of the BLA phase was referred to as CARVYKTI (the commercial product or product that was marketed). Successful manufacturing at 100% success was noted for all patients receiving ciltacabtagene autoleucel, the product that met specifications for the CARTITUDE-1 study. There were limitations in assessing the manufacturing failure rates for CARVYKTI. Samples for patients who did not receive ciltacabtagene autoleucel for clinical reasons of progressive disease, death or withdrawal from CARTITUDE-1 study were not available to evaluate for the post-hoc product release specifications for CARVYKTI. For these reasons the manufacturing failure rates for CARVYKTI (the to be marketed product) could not be assessed in all 113 leukapheresed patients but was assessed in the 97 patients who were considered efficacy evaluable by virtue of having received ciltacabtagene autoleucel.

Conclusion

The clinical development program for ciltacabtagene autoleucel in the indicated population beginning with the submission of an IND to a marketing approval was less than 4 years. The efficacy results from the CARTITUDE-1 trial provide evidence of clinical benefit of ciltacabtagene autoleucel in patients with relapsed or refractory multiple myeloma who have received at least 4 prior lines of therapy with previous exposure to a proteasome inhibitor, immunomodulatory agent and anti-CD38 antibody therapy. Key safety issues include CRS, HLH/MAS, neurologic toxicity including Parkinsonism, and prolonged cytopenia. The magnitude and durability of response, and rigorous risk mitigation measures support an overall favorable benefit-risk profile in the indicated patient population with limited therapeutic options (Refer to Table 4).⁵ Ciltacabtagene autoleucel represents another therapeutic option for patients with relapsed and refractory MM.

Table 4:

Risk and benefit considerations in ciltacabtagene autoleucel’s approval:

Decision Factor	Evidence and Uncertainties	Conclusions and Reasons
Analysis of Condition	Multiple myeloma (MM) is the second most common hematologic malignancy and accounts for 1.8% of all cancers and 17% of all hematologic malignancies. Therapy for patients with relapsed or refractory myeloma has improved considerably over the past three years with approval of multiple new therapies with improvement in response rate and progression free survival. However, relapsed, and refractory myeloma remains incurable, with a 5-year survival rate of 52%.	Relapsed or refractory multiple myeloma is a serious and life-threatening condition with need for effective and safe salvage therapies.
Unmet Medical Need	Patients with relapsed or refractory myeloma have unmet medical need.	Patient with relapsed or refractory myeloma have unmet medical need.
Clinical Benefit	In this single-arm multicenter study for patients with relapsed and refractory myeloma, lymphodepleting chemotherapy followed by ciltacabtagene autoleucel administered at dose range of 0.5 to 1.0 x 106 CAR+ T cells per kg body weight produced: Stringent CR rate of 78.4% (95% CI:68.8%,86.1%) according to IMWG 2016 criteria. ORR for the efficacy evaluable population, by independent review committee (IRC) assessment, of 97.9% (95% CI: 92.7%, 99.7%) with median duration of response of 21.8 months (95% CI:21.8, NE).	Based on the ORR, CR rate and DOR, ciltacabtagene autoleucel at the recommended dose range has clinically meaningful benefit in relapsed and refractory myeloma who have received a proteasome inhibitor, an IMiD agent, and an anti-CD38 antibody therapy.
Risk	Major AEs associated with ciltacabtagene autoleucel were cytokine release syndrome, neurologic toxicities, prolonged cytopenias; with some cases requiring stem cell rescue , infectious complications, hemophagocytic lymphohistiocytosis/macrophage activation syndrome (HLH/MAS), and hypogammaglobulinemia.	Available evidence indicates that the risk of ciltacabtagene autoleucel while substantial, does not outweigh the benefit to adult patients with relapsed and refractory myeloma.
Risk Management	The most substantial risks of ciltacabtagene autoleucel are CRS, neurologic toxicity (NT; includes ICANS, parkinsonism, GBS, peripheral neuropathy, cranial nerve palsy), HLH/MAS, prolonged and recurrent cytopenias, infections and persistent hypogammaglobulinemia. CRS and NT were mitigated in the trial by careful site selection and training of investigators. There are theoretical risks of secondary malignancy with this genetically modified immunotherapy based on the potential for replication competent lentivirus due to the risk of insertional mutagenesis.	The risks associated with ciltacabtagene autoleucel warrant boxed warnings - a REMS particularly for CRS,NT HLH/MAS and prolonged cytopenia requiring stem cell rescue therapy, and a long term follow up study for risk assessment of subsequent malignancy attributable to insertional mutagenesis.

Data from Source: BLA clinical review memorandum 125746/0⁵