EHA 2017 | Treatment of relapsed/refractory aggressive NHL 

On Saturday 24^th June, during the 22nd Congress of the European Hematology Association (EHA), a session took place discussing therapy for aggressive relapsed/refractory Non-Hodgkin Lymphoma (NHL). The session was co-chaired by Dolores Caballero from the University Hospital of Salamanca, Salamanca, Spain, and Martin Hutchings from Copenhagen University Hospital, Copenhagen, Denmark. Below are the key highlights from three of the five presentations that made up this session.

Abstract S466: ZUMA-1 results from the primary analysis

The first abstract was presented by Yi Lin from the Mayo Clinic, Rochester, United States, and was on the clinical and biologic covariates of outcome from the primary analysis of the phase II ZUMA-1 trial (NCT02348216) of the CAR-T construct KTE-C19 (Axi-cel) in the treatment of refractory DLBCL, PMBCL and TFL. The data presented was after a median follow-up of 8.7 months (data were also presented at ICML 2017, you can find our coverage here).

Summary:

• Refractory Aggressive NHL currently has a poor outcome with ORR = 26% (CR = 8%, PR = 18%), demonstrating a high unmet clinical need, based on the retrospective SCHOLAR-1 meta-analysis
• Treatment: low-dose conditioning therapy (cyclophosphamide 500 mg/m² and fludarabine 30 mg/m² for three days) followed by weight-based infusion of 2x10⁶ CAR-T cells/kg (successful manufacturing rate of 99%)
• 111 pts enrolled, 101 received KTE-C19 and are included in the modified intention to treat analysis

Results:

• Six-month OS = 80%, median OS = not reached
• DLBCL (n = 77): ORR = 82%, CR = 49%
• PMBCL or TFL (n = 24): ORR = 83%, CR = 71%
• Total (n = 101): ORR = 83%, CR = 54%
• At 8.7 months follow-up: ORR = 44%

It was observed that PFS rates were consistent across key covariates including age and IPI score, and that response rates were also consistent between ABC and GCB cell of origin subtypes of DLBCL with an ORR of 76% and 88% respectively. The most frequent grade 3 or higher treatment-emergent adverse events were anemia (43%), neutropenia (39%), and reduced neutrophil counts (32%). Importantly, febrile neutropenia was observed in 31% of patients, and of particular interest related to CAR-T therapy was the rate of encephalopathy at 21%. Additionally, cytokine release syndrome and neurological events were reported to be ‘generally reversible’. Three patients died during the interim analysis period, with no new treatment-related deaths reported in the primary analysis. Also relevant for this type of therapy is the number of patients which received steroids (27%) or tocilizumab (43%) to help manage the AEs. Interestingly, use of these agents was found to be associated with higher levels of CAR T-cell expansion, which in turn was associated with higher ORR and grade 3 AE rates. Further data from the ZUMA-1 trial were also presented at ASH 2016 (here), and at two presentations during AACR 2017.

Abstract S467: Combination of CC-122 with obinutuzumab

The next abstract was presented by Reda Bouabdallah from the Institut Paoli Calmettes, Marseille, France, which reported the results from the phase Ib study of the combination of CC-122 and obinutuzumab in the treatment of relapsed/refractory B-cell NHL (NCT02417285).

CC-122 is an agent which binds to cereblon, and increases the degradation of Aiolos and Ikaros resulting in immune modulation and anti-lymphoma activities. The combination of CC-122 with the anti-CD20 antibody obinutuzumab has been shown to result in a synergistic increase in anti-lymphoma effects in DLBCL in pre-clinical trials. 

This phase Ib trial used the study design shown below. The primary endpoint was the safety and tolerability of the combination therapy, along with identifying the maximum tolerated dose, non-tolerated dose and recommended phase II dose for CC-122 in combination with obinutuzumab. Secondary endpoints were ORR, DOR and PFS.

Summary:

• Total number of pts recruited = 38
• Male = 68%, median number of prior therapies = 4, prior SCT = 37%
• Results:
  • Total ORR = 66%, CR = 32%, PR = 34%
  • DLBCL ORR = 47%, CR = 16%, PR = 32%
  • FL/MZL OFF = 84%, CR = 47%, PR = 37%
  • Pts receiving 3 mg or higher of CC-122 + obinutuzumab have shown higher and more durable responses
  • Therefore, 3 mg was concluded to be the recommended phase II dose
• Adverse events are detailed below:

In conclusion, Dr Bouabdallah stated that CC-122 in combination with obinutuzumab was shown to be clinically meaningful in this heavily pre-treated population and that the combination was well-tolerated with grade 3 or higher AEs being mostly myelosuppression which was manageable.

Abstract S468: Polatuzumab vedotin + BR or BG – Updated Phase Ib/II study

The third abstract of this session was presented by Matthew Matasar from the Memorial Sloan Kettering Cancer Center, New York, US, on the updated data from the phase Ib/II GO29365 trial of polatuzumab vedotin with either bendamustine + rituximab (BR) or bendamustine + obinutuzumab (BG) (NCT02257567).

Polatuzumab vedotin (Pola) is an antibody-drug conjugate (ADC) which is comprised of a CD79b monoclonal antibody to target malignant B-cells, connected via a peptide linker to a microtubule-disrupting agent, monomethyl auristatin E (MMAE). Once the ADC reaches the lysosome inside malignant cells MMAE is released, which results in microtubule disruption and apoptosis.

The primary endpoint of the Phase Ib component was safety and tolerability of the combination, and to recommend a dose for the phase II study. The secondary endpoints included CR as measured by PET, OR based on PET or CT, and best objective response.

In the phase I part of the study, 6 FL and 6 DLBCL patients were given Pola + BR, and in the phase Ib/II part 26 FL and 26 DLBCL patients were given Pola + BG. Patients had received a median of two lines of prior treatment.

Summary:

• Phase Ib (Pola + BR)
  • FL (n = 6): ORR = 100%, CR = 67%, PR = 33% (18.7-months median follow-up)
    • Median PFS = 18.4 months
  • DLBCL (N = 6): ORR = 50%, CR = 33%, PR = 17% (20-months median follow-up)
    • Median PFS = not reached
  • Phase Ib/II (Pola + BG)
    • FL (n = 26): ORR = 89%, CR = 65%, PR = 23% (10.2-months median follow-up)
    • DLBCL (N = 26): ORR = 60%, CR = 41%, PR = 19% (9.3-months median follow-up)
  • Safety:
    • Cytopenias were the most common reason for dose reduction or delay of polatuzumab vedotin, with one discontinuation
    • 33% pts has peripheral neuropathy, most were grade 1, however this lead to dose changes or discontinuation in 8% of pts
    • Incidence of hematological adverse events grade 3 or 4 were higher in DLBCL pts than FL pts
    • Any serious adverse events occurring in over 10% of pts were observed in 69% of DLBCL pts and 41% of FL pts
    • In total, 7 grade 5 AEs were reported, 2 in FL Pola + BG group, and 5 in DLBCL Pola + BG group.

In conclusion, the phase Ib/II portion of this study into the combination of polatuzumab plus bendamustine with either rituximab or obinutuzumab reported in R/R FL patients a combined ORR of 91% (CR = 66%), and a combined ORR of 58% (CR = 39%) in R/R DLBCL patients, with some durable responses reported after a year. The phase II portion of the study comparing Pola + BR vs. BR alone is continuing.