The effect of inotuzumab ozogamicin treatment exposure and cycles on R/R ALL outcomes

Inotuzumab ozogamicin (InO) is an anti‐CD22 monoclonal antibody conjugated to calicheamicin that has been licensed by the European Medicines Agency (EMA) and the U.S. Food and Drug Administration (FDA) for the treatment of adults with relapsed or refractory (R/R) acute lymphoblastic leukemia (ALL).^1,2 The approvals were based on the phase I–II trial 1010 (NCT01363297) and the phase III INO‐VATE trial (NCT01564784). In the INO-VATE trial, patients randomized to InO achieved higher complete response/complete response with incomplete hematologic recovery rates (CR/CRi; 74% vs 35%) and measurable residual disease (MRD)-negativity rates (62% vs 17%), when compared to patients receiving chemotherapy by investigator’s choice and similar results were seen in patients receiving InO treated in study 1010.¹

Recently, two sub-analyses of the INO-VATE and/or study 1010 trials investigated the effect of InO exposure and number of treatment cycles on safety and efficacy outcomes for patients with R/R ALL.^1,2 Study A, which evaluated InO exposure, was published by Joseph Chen et al. in Clinical and Translational Science,¹ while study B, investigating InO treatment cycle number, was published by Ryan D. Cassaday et al. in the British Journal of Haematology.² We hereby summarize the results of both analyses.

Study A — Analysing InO exposure¹

Study design

• All included patients were diagnosed with R/R CD22‐positive B‐cell ALL.
• Data were pooled from the 1010 and INO-VATE trials for the InO arm and from the INO-VATE trial alone for the control arm (investigator’s choice chemotherapy).
  • The pooled population for the exposure–response analysis included patients who had available pharmacokinetics (PK) and had received at least one dose of InO (study 1010, n = 72; INO-VATE, n = 162).
• PK parameters were measured with a previously published InO population PK model.³ The measured InO exposure PK parameters were:
  • C_maxevent: Maximum observed concentration prior to response
  • C_maxoverall: Maximum observed concentration for the duration of treatment
  • C_AUC: Cumulative area under the concentration-time curve
  • C_avg: Average plasma concentration
  • C_AUCP1: C_AUC in the first treatment cycle
• The following safety end points were considered in the exposure-response analysis:
  • Hepatic events
  • Investigator-reported veno‐occlusive disease/sinusoidal obstruction syndrome (VOD/SOS)
  • Hepatic Event Adjudication Board (HEAB)‐assessed VOD/SOS
  • Grade ≥ 3 adverse events (neutropenia, thrombocytopenia, elevated bilirubin, aspartame aminotransferase, and alanine aminotransferase)
• Binomial logistical regression analysis was performed for modelling data interpretation.
• The key baseline patient characteristics are shown below in Table 1.

Table 1. Key baseline patient characteristics from Study A¹

Results

• Logistical regression analyses in the exposure–response population revealed the following CR/CRi predictors:
  • C_avg (p < 0.0001)
  • The percentage of leukemic blasts that were CD22-positive at baseline measured either from bone marrow or peripheral blood samples (p = 0.0007)
• Logistical regression analyses in the exposure–response population revealed the following MRD-negativity predictors:
  • C_avg (p < 0.0001)
  • The percentage of leukemic blasts that were CD22-positive at baseline measured either from bone marrow or peripheral blood samples (p = 0.0001)
  • Cytogenetics at baseline (p = 0.0067)
• Logistical regression analyses of safety endpoints revealed that InO exposure has a significant (positive) association only with HEAB-assessed VOD/SOS.
• Moreover, post-study hematopoietic stem cell transplantation (HSCT) was found to be a statistically significant predictor of HEAB-assessed VOD/SOS (p = 0.0026).
• The odds of achieving CR/CRi in the InO treatment arm, with a given baseline ECOG PS and amount of blasts, was estimated at 7.1-times higher than in the control arm (OR, 7.1; 95% CI, 4.3–12.3).
• The odds of achieving MRD-negativity in the InO arm with a given baseline ECOG PS, number of peripheral blasts, and given cytogenetics, with or without prior HSCT, were estimated at 12.7-times higher than the control arm (OR, 12.7; 95% CI, 7.0–24.3).

Study B — Analyzing InO treatment cycles²

Study design

• Post-HSCT efficacy outcomes were evaluated in respect to the number of InO treatment cycles received.
• Data were extrapolated from the INO-VATE trial dataset and only patients achieving CR/CRi were analyzed.
• The key baseline characteristics for the patients who proceeded directly to HSCT following InO treatment are shown below in Table 2.

Table 2. Key baseline patient characteristics from Study B²

Results

• In total, 71 patients achieved CR/CRi and underwent HSCT. Of those, 65 patients proceeded directly to HSCT after InO treatment without receiving salvage therapy in between. Fifty patients achieved CR and did not proceed to HSCT.
• All treatment groups had higher chances of achieving MRD negativity with increasing number of InO treatment cycles.
• For patients who underwent HSCT directly after InO treatment, non‐relapse mortality rate was lower, progression-free survival (PFS), post‐HSCT survival, and overall survival (OS) were longer when only ≤ 2 InO cycles were administered (Table 3). Also, in terms of safety, post‐HSCT VOD/SOS occurred only in 17.1% of patients receiving ≤ 2 InO cycles compared to 33.3% receiving > 2 InO cycles.

• In the 50 patients that did not proceed to HSCT following InO treatment, PFS and OS were significantly longer in those having received > 3 InO cycles than those with ≤ 3 cycles. Moreover, the rate of serious treatment‐emergent adverse events (59.4% vs 66.7%) and the percentage of patients discontinuing due to adverse events (28.1% vs 38.9%) were lower in the > 3 vs ≤ 3 cycles InO cycles groups, respectively.

Table 3. Efficacy outcomes from Study B of patients who achieved CR/CRi and proceeded directly to HSCT after Ino²

Conclusions

Both studies indicate that higher exposure to InO is associated with better efficacy outcomes, in terms of achieving higher CR/CRi and MRD negativity confirming the recommended dosing of InO with 1.8 mg/m² per cycle. When analyzing the optimal number of treatment cycles, study B showed that the best survival outcomes were achieved for patients not proceeding to HSCT when InO was given for more than 3 (and up to 6) cycles. In contrast, patients proceeding to HSCT benefited most from ≤ 2 cycles of InO treatment when directly followed by HSCT without conditioning.

However, InO treatment is associated with a risk for VOD/SOS, which seems to increase with higher exposure and number of cycles. This requires treating physicians to carefully weigh the benefits and risks and to consider dose reduction in patients at higher risk for VOD/SOS. It should be noted that both studies are retrospective post-hoc analyses with relatively small sample sizes for their explorative nature. Thus, the results presented above require further clinical validation from large-scale prospective studies.