ASCO 2019 | COLUMBA trial: intravenous versus subcutaneous daratumumab

On Sunday, 2 June, 2019, during the annual American Society of Clinical Oncology (ASCO) meeting in Chicago, US, our Steering Committee member, Maria-Victoria Mateos from University of Salamanca, Salamanca, ES, presented data from the COLUMBA trial (NCT03277105).¹

COLUMBA is a phase III trial designed to show that the administration of daratumumab subcutaneously (SC) is non-inferior to the currently used, intravenous (IV) method, for the treatment of relapsed/refractory multiple myeloma (RRMM).

Background

The first IV administration of daratumumab currently takes, on average, 7.0 hours, followed by 4.3 hours and 3.4 hours for the second and third doses respectively. Co-formulating daratumumab with recombinant human hyaluronidase PH20 (rHuPH20), increases the permeability of SC tissue, creating a SC option (dara-SC). Prior studies (Ib PAVO study, NCT02519452) have shown SC daratumumab at 1,800mg increases efficacy and pharmacokinetics (PK), and lowers the rate of infusion related reactions (IRRs) when compared to intravenous daratumumab (dara-IV).²

Data given as dara-IV versus dara-SC unless otherwise stated

Study design and patient characteristics

• Patient eligibility (N= 522)
  • RRMM having received ≥3 prior lines of therapy
  • Prior therapy: proteasome inhibitor (PI) and an immunomodulatory drug (IMiD)
  • Refractory to a PI and an IMiD
  • Patient characteristics:
    • Median age (years): 68 (33–92) vs 65 (42–84)
    • Patients ≥75 years old: 23% vs 18%
    • ECOG status ≥2: 15% vs 18%
• Randomization 1:1 to either;
  • Dara-IV: 16mg/kg (n= 259)
  • Dara-SC: 1800mg (n= 263)
  • Stratified by body weight, prior lines of therapy and myeloma subtype
• Both were given weekly (QW) in cycles 1–2, twice weekly (Q2W) in cycles 3–6 and monthly (Q4W) in cycle 7 onwards, until progressive disease (PD)
• Co-primary endpoints; overall response rate (ORR) and maximum C_trough
  • ORR: non-inferiority or superiority of dara-SC would be shown if the following criteria were met:
    • Non-inferiority: lower bound of 95% confidence interval (CI) of ≥60%
    • Superiority: lower bound of 95% CI of relative risk of ≥100%
  • Maximum C_trough:
    • Non-inferiority: lower bound of 95% CI of geometric means of C_trough on cycle 3, day 1 (C3D1), of ≥80%
• Secondary endpoints included; IRR rate, progression-free survival (PFS), rate of very good partial response (VGPR) or better and complete response (CR) or better, time to next therapy, overall survival (OS) and patient reported outcomes (PROs)
• Median follow-up: 7.46 months (0.03–13.86)

Safety

• The rate of discontinuation was the same in both arms (57%) with the main reasons being PD or AEs
• Median duration of infusions are shown in Table 1, with dara-SC significantly reducing the administration time compared to dara-IV
  • Median duration of treatment: 5 months
  • Median number of cycles: 6
  • Median relative dose intensity for both groups: 100%

Table 1: Median duration of infusion per administration, by method of administration

Dara, daratumumab; IV, intravenous; SC, subcutaneous
	Median duration of infusion (minutes)
First
Dara-IV	421 (7.0 hours)
Dara-SC	5
Second
Dara-IV	255 (4.3 hours)
Dara-SC	5
Subsequent
Dara-IV	205 (3.4 hours)
Dara-SC	5

• IRR: 34.5% vs 12.7%
  • Odds ratio (OR), 95% CI: 0.28 (0.18–0.44), P < 0.0001
  • Most were mild and occurred at the first administration
  • Grade 3 IRR: 5.4% vs 1.5%
  • No grade 4 events
  • Median time to onset: 1.5 hours vs 3.6 hours
  • Injection-site reactions in dara-SC group: 6.9%
    • Grade 1 or 2 only
• Common AEs are shown in Table 2
  • Treatment-emergent AEs (TEAEs) grade ≥3: 49% vs 46%
  • TEAEs causing patient to discontinue treatment: 8% vs 7%
  • Grade 5 TEAE: 7% vs 5%
• PROs: Based on a modified cancer treatment questionnaire, patients receiving dara-SC were more satisfied than dara-IV patients

Table 2: Treatment-emergent AEs (TEAEs) grade ≥3 (hematologic TEAEs occurring in ≥5% of patients, and non-hematologic TEAEs in ≥2% of patients)

	Dara-IV (n = 258)	Dara-SC (n = 260)
Hematologic (%)
Anemia	14%	13%
Thrombocytopenia	14%	14%
Neutropenia	8%	13%
Lymphopenia	6%	5%
Non-hematologic (%)
Back-pain	3%	2%
Hypertension	6%	3%
Dara, daratumumab; IV, intravenous; SC, subcutaneous

Efficacy: Co-primary endpoints

• Response data is shown in Table 3
• Relative risk (95% CI): 1.11 (0.89–1.37), P < 0.0001
• Dara-SC is non-inferior to dara-IV with comparable ≥VGPR and ≥CR rates
• Subgroup analysis: ORR was comparable, including by body weight

Table 3: Response rates by method of administration of daratumumab

CR, complete response; Dara, daratumumab; IV, intravenous; ORR, overall response rate; PR, partial response; SC, subcutaneous; VGPR, very good partial response
	Dara-IV	Dara-SC
ORR	37.1%	41.1%
≥CR	2.7%	1.9%
≥VGPR	17.0%	19.0%
PR	20.1%	22.1%

•  Maximum C_trough concentration at C3D1 ratio, dara-SC/dara-IV: 107.93%
  • At a 90% CI the range was 95.74%–121.67%
  • Lower bound of 95.74% meaning dara-SC met the PK non-inferiority criteria of above 80%

Efficacy: Secondary endpoints

Median PFS and OS was similar between groups, with the data shown in Table 4.

Table 4: Median PFS and OS data in the ITT population

Dara, daratumumab; HR, hazard ratio; IV, intravenous; PFS, progression free survival; OS, overall survival; SC, subcutaneous
	Median PFS	6-month OS rate
Dara-IV	6.1 months (4.7–8.3)	83% (77.6–87.2)
Dara-SC	5.6 months (4.7–7.6)	87.5% (82.7–91.0)
HR, 95% CI	0.99 (0.78–1.26)	0.90 (0.59–1.35)
P value	0.9258	0.6032

Conclusion

Maria-Victoria Mateos, and co-authors from the study, concluded that dara-SC is non-inferior to dara-IV as the study met both co-primary endpoints. The safety profile of both is comparable, with dara-SC being associated with a lower rate of IRRs. Significantly, patients reported an improved experience with dara-SC, based on the shorter administration time of 5 minutes, compared to hours for dara-IV.

This study supports the use of flat-dose 1,800mg dara-SC in RRMM patients having received ≥3 prior lines of therapy.

UPDATE: The results of the study were published in Lancet Haematology in March 2020 and can be accessed here