Results from the ALPHA study on ALLO-501 and lymphodepleting ALLO-647 in R/R DLBCL or FL

Despite the success of chimeric antigen receptor (CAR) T-cell therapy in treating hematologic malignancies, many limitations still exist, including the cost, the time between leukapheresis and infusion, and the quality of T cells.¹

Strategies to overcome these limitations include the use of universal or allogeneic donor “off-the-shelf” CAR T cells produced from healthy donor peripheral blood mononuclear cells.¹

The ALPHA study is a first-in-human phase I trial (NCT03939026) of ALLO-501 and ALLO-647 for relapsed/refractory (R/R) diffuse large B cell lymphoma (DLBCL) or follicular lymphoma (FL). ALLO-501 is an allogeneic CAR T-cell therapy product that is generated by introducing an anti-CD19 CAR molecule into healthy donor T cells.²

One of the major obstacles to the success of the allogeneic approach is the risk of graft-versus-host disease (GvHD). To minimize the risk of GvHD in ALLO-501, the T-cell receptor α was knocked out using the TALEN gene-editing technology. In addition, to prevent host rejection and to improve persistence, CD52 was knocked out to allow for selective lymphodepletion with ALLO-647, an anti-CD52 monoclonal antibody that depletes host T cells but protects donor cells that do not express CD52, allowing their expansion and persistence.²

Study design²

This phase I, open-label, multicenter, dose-escalation study evaluated patients with either R/R DLBCL or FL after at least two prior lines of therapy, including an anti-CD20 monoclonal antibody; prior autologous CAR T-cell therapy was allowed if the tumors remained CD19+.

The primary endpoint was safety and dose-limiting toxicity, and secondary endpoints included overall response rate (ORR), ALLO-501 cell kinetics, and ALLO-647 pharmacokinetics.

Treatment schedule²

After enrollment, patients received lymphodepletion starting on Day −5 or Day −7 with one of the following regimens until infusion:

• LD1: fludarabine (Flu) 30 mg/m²/day × 3 days, cyclophosphamide (Cy) 300 mg/m²/day × 3 days, and ALLO-647 13 mg/day × 3 days (39 mg in total)
• LD2/LD3: Flu/Cy as LD1 and ALLO-647 30 mg/day × 3 days (90 mg in total), either concomitantly or in a staggered manner

Lymphodepletion was followed by a single infusion of ALLO-501 on Day 0. The first tumor assessment was performed at Day 28.

Three CAR T-cell doses were tested: 40 × 10⁶ (DL1), 120 × 10⁶ (DL2), and 360 × 10⁶ (DL3).

Patient characteristics²

The baseline patient characteristics are reported in Table 1. Patients were heavily pre-treated with a median number of prior regimens of four and advanced-stage disease.

Table 1. Patient characteristics²

CAR, chimeric antigen receptor; DL, dose level; FLIPI, Follicular Lymphoma International Prognostic Index; HSCT, hematopoietic stem cell transplant *One patient with stage II disease was treated at DL1
Characteristic	DL1 (n = 4)	DL2 (n = 10)	DL3 (n = 8)	All patients (N = 22)
Median age, years (range)	57 (42─67)	70 (37─73)	54 (34─67)	63 (34─73)
Male, n (%)	3 (75)	8 (80)	6 (75)	17 (77)
Lymphoma subtypes, n (%)
Diffuse large B-cell lymphoma	3 (75)	5 (50)	6 (75)	14 (64)
Follicular lymphoma	1 (25)	5 (50)	2 (25)	8 (36)
Disease stage (per Lugano 2014), n (%)*
Stage III	1 (25)	5 (50)	2 (25)	8 (36)
Stage IV	2 (50)	5 (50)	6 (75)	13 (59)
FLIPI score 3─5, n (%)	1 (25)	6 (60)	5 (63)	12 (55)
Prior treatments
Median number (range)	2 (2─4)	4 (3─4)	5 (3─8)	4 (2─8)
HSCT, n (%)	2 (50)	4 (40)	3 (38)	9 (41)
Autologous CAR T-cell, n (%)	—	1 (10)	3 (38)	4 (18)

Results²

At the time of data cut-off, on May 11, 2020, 22 patients were available for safety, and 19 patients were available for efficacy. The median time from enrollment to start of lymphodepletion was short (5 days) reflecting the potential advantage of allogeneic CAR-T-cell therapy over autologous CAR T-cell therapy. All patients treated to date received CAR-T cells generated from a single healthy donor. The dose escalation has been completed for the CAR T-cells at the low dose ALLO-647 (n = 11) and is ongoing on the high dose ALLO-647 (n = 11).

Safety²

Both ALLO-501 and ALLO-647 demonstrated a manageable safety profile with no dose-limiting toxicities, GvHD, or neurological toxicty. Adverse events of interest are reported in Table 2.

Table 2. Adverse events of interest²

CRS, cytokine release syndrome *Attributed to ALLO-647
AE of interest	Grade 1 n (%)	Grade 2 n (%)	Grade 3 n (%)	Grade 4 n (%)	All grades n (%)
CRS	2 (9)	4 (18)	1 (5)	—	7 (32)
Infection	5 (32)	4 (18)	2 (9)	—	11 (50)
Infusion reaction*	1 (9)	9 (41)	1 (9)	—	11 (50)
Neutropenia	—	1 (5)	7 (32)	7 (32)	15 (68)

All the observed cytokine release syndrome (CRS) events were manageable.

The median duration of hospitalization from infusion was 7 days.

Serious adverse events occurred in 18% of patients (n = 4), and included Grade 2 pyrexia and cytomegalovirus (CMV) reactivation, Grade 3 rotavirus infection and hypokalemia, Grade 3 febrile neutropenia and hypotension, as well as Grade 3 upper gastrointestinal hemorrhage and CMV reactivation. All of these were eventually resolved.

Efficacy²

After a median follow-up of 3.8 months (range, 0.7─6.1), the  ORR was 63% (95% CI, 38─84; n = 12), and the best complete response (CR) rate was 37% (95% CI, 16─62; n = 7), among the 19 patients available for efficacy analysis. The CR rate, but not the ORR, was higher with the higher ALLO-647 lymphodepletion dose compared with the lower dose (50% vs 27%). The median time to response was 1 month, and nine of the 12 responses are ongoing.

Ten out of 16 patients, evaluated to date, had evidence of CAR T-cell expansion, with patients in CR achieving a higher peak and persistence (observed up to Day 56) compared with patients with stable and progressive disease.

ALLO-647 mediated a selective lymphodepletion with a deeper T-cell lymphodepletion and delayed T cell recovery at the higher ALLO-647 dose (90 mg) compared with the lower dose (39 mg). No difference in neutrophil kinetics was observed between the two treatment groups.

Conclusion

ALLO-501 and ALLO-647-based lymphodepletion were well tolerated, with no dose-limiting toxicities, GvHD, or neurotoxicity. The observed CRS events were manageable. The short-term preliminary efficacy was comparable to autologous CAR-T products that are currently in clinic, with an ORR of 63%. Further optimization of the lymphodepleting regimen and a longer patient follow-up to determine response durability are currently ongoing.