ASCO 2019 | AMG 420 first-in-human phase I dose escalation study results

On Sunday 2 June, 2019, Professor Max Topp, University of Würzburg, Würzburg, DE, presented the results of the phase I trial of AMG 420 (BI 836909) in relapsed/refractory multiple myeloma (RRMM), at the annual American Society of Clinical Oncology (ASCO) meeting, in Chicago, US.

Background

AMG 420 is a bispecific T-cell engager (BiTE®) antibody construct that binds to B-cell maturation antigen (BCMA) on tumor/plasma cells, and CD3 on T-cells. This dual action leads to T-cell mediated lysis of BCMA⁺ cells.

Study design and patient characteristics (NCT02514239)

• Patients with RRMM who have received 2 or more prior lines of therapy including ≥1 proteasome inhibitor (PI) and ≥1 immunomodulatory drug (IMiD)
  • Prior allogeneic stem cell transplant was an exclusion criteria
  • Characteristics shown in Table 1
• Endpoints:
  • Primary: Maximum tolerated dose (MTD) and dose-limiting toxicities (DLTs)
  • Secondary: Response including minimal residual disease (MRD) negativity defined as <1 tumor cell per 10⁴ normal cells in bone marrow by flow cytometry
• AMG 420 administered by continuous infusion in 6-week cycles for up to 10 cycles
  • Initial single patient cohorts dosed at: 0.2–1.6μg/day
  • Subsequent, three to six patient cohorts dosed at: 3.2–800μg/day
• Given for ≤5 cycles or until disease progression (PD), toxicity or consent being withdrawn
• Five further cycles, up to a total of 10, could be given dependent upon response:
  • Cycle 1: 4 weeks of AMG 420
    • ≤4 days in clinic
  • Cycle 2: 4 weeks of AMG 420
    • ≤24 hours in clinic
  • Cycle 3 onwards (up to 10): 4 weeks of AMG 420
    • ≤8 hours in clinic
  • There was a treatment-free period of 2 weeks between cycles
  • Safety follow-up visit 30-days after end of treatment

Table 1: Patient characteristics

ECOG, Eastern Cooperative Oncology Group; IMid, immunomodulatory drugs; PI, proteosome inhibitor
	N= 42
Median age (year)	65 (39–79)
ECOG status (0 vs 1 vs 2)	57% vs 40% vs 2%
Cytogenetics (standard vs intermediate vs high)	55% vs 31% vs 14%
Median prior lines of therapy	4 (2–13)
Prior monoclonal antibody (daratumumab vs elotuzumab)	26% vs 10%
Refractory to monoclonal antibody (daratumumab vs elotuzumab)	21% vs 10%
Refractory to IMid/PI/IMid and PI	55% vs 45% vs 31%

Primary endpoints: DLTs and MTD

Table 2 shows a summary of the three grade 3 DLTs that occurred. The MTD was determined to be 400μg/day. No anti-AMG 420 antibodies were detected up to 800μg/day.

Table 2: Summary of grade 3 DLTs

	DLT	Dose of AMG 420	Timing of DLT	Notes
1	CRS	800μg/day	1 day after treatment start. Resolved after AMG 420 was discontinued	Symptoms; fever, hypertension, tachycardia and retrograde amnesia
2	Peripheral PN	800μg/day	1 week after treatment start, required hospitalization by day 14	M protein decreased by 58% after 15 days of treatment. PN resolved with IV Ig and corticosteroid
3	PN	400μg/day	Grade 1 PN developed to grade 3 by day 30 and improved to baseline by day 84	M protein decreased by 79% after 28 days of treatment. PN resolved with IV Ig and corticosteroid
CRS, cytokine relase syndrome; DLT, dose-limiting toxicities; PN, polyneuropathy

CRS adverse events (AEs) and serious AEs (SAEs)

The safety data for CRS, SAEs and treatment-related SAEs is shown in Table 3. Of the two patients who died of infection, one died of aspergillosis/flu and one died of fulminant hepatitis related to adenovirus infection. Neither were deemed to be treatment-related.

Of the 19 patients (45%) experiencing a SAE, 16 were hospitalized, 4 for a prolonged period. No grade 3–4 central nervous system toxicities were observed.

Table 3: CRS, SAEs and treatment-related SAEs grade 1–5 (N= 42)

CRS, cytokine relase syndrome; PN, polyneuropathy; SAEs, serious adverse events
Grade	Total	1	2	3	4	5
CRS	16 (38%)	13	2	1	-	-
SAEs (≥2 patients)
Infections	13 (31%)	-	3	8	-	2
PN	2 (5%)	-	-	2	-	-
Treatment-related SAEs
PN	2 (5%)	-	-	2	-	-
Edema	1 (2%)	-	-	1	-	-

Discontinuations

Currently, two patients remain on therapy, with the remainder having discontinued for the following reasons; PD (n= 25, 60%), AEs (n= 7, 17%), death (n= 4, 10%), completion of 10 cycles (n= 3, 7%), and withdrawal of consent (n= 1, 2%). Deaths were attributable to disease progression (n= 2) and AEs (n= 2).

Efficacy

At a dose of 400μg/day (n= 10), 5 MRD negative complete responses (CRs) were seen, with 1 very good partial response (VGPR) and 1 partial response (PR) giving a response rate of 70%, lasting for a median of 9.0 months (5.8–13.6).

In total, 13 patients responded (13/42), with a median time to response of 1 month, lasting for 8.4 months on average (2.5–15.5). MRD negative CRs lasted for a median of 9.6 months (2.8–12.8), with 5 of these patients in the 400μg/day cohort. Aside from the patients whose last response at evaluation was PD (n= 5), all patients were responding.

Responses at doses less than 200μg/day were:

• Three patients responded at doses of:
  • 5μg/day, completed full 10 cycles
    • Best response: CR in cycle 8
  • 50μg/day, completed full 10 cycles
    • Best response: PR in cycles 2–5
  • 100μg/day, discontinued due to PD having received 8 cycles
    • Best response: CR in cycles 4–5
• Responses at doses ≥200μg/day are shown in Table 4

Table 4: Characteristics of responding patients at doses ≥200μg/day

CR, complete response; EOT; end of treatment, N/A; patient received 10 cycles or treatment ongoing as of April 2019 (maximum); MRD, minimum measurable disease; PD, progressive disease; PN, polyneuropathy; PR, partial response; VGPR, very good partial response
Dose (μg/day)	Number of cycles to response	Prior lines of therapy	Reason for discontinuation	Best response	Cycle of best response
200	4	6	Port infection	MRD- CR	3–4 to 11 months post-treatment
400	7	3	N/A	MRD- CR	3–7
400	10	4	N/A	MRD- CR	3–10
400	8	6	PD	MRD- CR	1–7
400	10	2	N/A	MRD- CR	1–10
400	5	5	PD	MRD- CR	3–4
400	1	5	PN	VGPR	From EOT to 7 months post-EOT
400	1	5	Death	PR	1
800 400 (dose reduction)	2 1	5	Withdrew consent	VGPR	2–3
800	0.5	5	PN	PR and CR	PR in 1, CR 9 months post-EOT

Conclusion

AMG 420 demonstrates activity in a heavily pre-treated population with RRMM who had received, on average, 4 prior lines of therapy. At the MTD of 400μg/day there was a 70% response rate, with 5 patients achieving MRD negative CR. No major toxicities were reported in the dose escalation portion of the study, subsequently DLTs were experienced by three patients. In addition, 13 (31%) patients experienced infections, 10 of which were grade 3 or above. Professor Topp suggested that future clinical trials of AMG 420 should thoroughly evaluate infection risk in order to best inform AE management guidelines. A phase Ib/II trial is ongoing.