Allogeneic Cell Therapy in Refractory T-Cell Lymphoma

Scientific background

Peripheral T-cell lymphoma (PTCL) is a rare group of diseases. There are nodal, extranodal and leukemic forms, the latter two predominating in Asia. The major nodal entities are anaplastic large cell lymphoma (ALCL), PTCL, not otherwise specified (NOS), and nodal T-follicular helper cell lymphoma (nTFHL) with its major subtype angioimmunoblastic T-cell lymphoma (AITL).

Except for anaplastic lymphoma kinase (ALK)-positive ALCL, the prognosis of nodal PTCL is poor, with 2-year progression-free survival (PFS) rates of 27-40%. The focus of the REACT study is nodal ALK-negative PTCL, here collectively referred to as high-risk nodal PTCL. The three entities PTCL-NOS, AITL, and ALK-negative ALCL constitute about half of all PTCL cases.

A particular feature of PTCL is primary refractory disease. Despite response to the first few treatment cycles, about a third of patients suffer progression on continued therapy, and among those without overt progression, a substantial number fail to reach a complete remission at the end of treatment (EoT). Because salvage protocols have limited anti-T-cell lymphoma activity, most primary refractory patients never achieve a remission and succumb to their disease. Patients relapsing after a longer period of disease control fare better, because allogeneic transplantation (alloHCT), the single most effective treatment modality, is curative in about half of the patients transplanted. The major hurdle preventing patients with primary refractory disease from undergoing alloHCT is progression, which may be overcome by identification of such patients before progression occurs. Toxicities from salvage therapies further reduce the number of patients able to receive an allogeneic transplant.

In the PTCL subgroup of the PETAL trial (ClinicalTrials.gov NCT00554164), a large multi-center lymphoma trial performed in Germany, 18Fluorine-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) identified PTCL patients with treatment failure before overt progression occurred. This finding may open a window of opportunity for alloHCT for a substantial proportion of patients with otherwise devastating prognosis. First-line alloHCT as consolidation after CHOEP induction therapy was pioneered in non-selected PTCL patients in the German-French AATT trial (ClinicalTrials.gov NCT00984412).

The novelty in the design of the REACT study is not in the treatment regimens, but in the use of interim PET/CT to perform alloHCT in chemotherapy-refractory patients early during first-line therapy. The underlying hypothesis is that early recognition of treatment failure will allow the affected patients to undergo alloHCT in good condition and at low tumor burden, which may improve survival for the entire high-risk nodal PTCL group.

Rationale for choice of control interventions/comparators

REACT is a non-randomized study without control intervention. The outcome of patients with iPET-defined treatment failure continuing chemotherapy is dismal. Medical ethics preclude random allocation of such patients to continued chemotherapy vs. alloHCT. The results of the REACT study will be compared with those of the PETAL trial where chemotherapy was continued irrespective of the iPET result, representing the current standard of care.

Risk-benefit assessment

About 40% of high-risk nodal PTCL patients experience chemotherapy-refractory disease in the course of first-line treatment, with less than 20% alive after 30 months. Treatment options for such patients are very limited. There are no promising drug treatments for chemotherapy-refractory high-risk nodal PTCL patients. AlloHCT offers a promising approach for patients with iPET-defined treatment failure that would until now be continued on chemotherapy. AlloHCT has a potent graft-versus-PTCL effect, it represents the only cellular immunotherapy option against PTCL. AlloHCT is associated with procedure-associated risks of infections and graft-versus-host disease, amounting to a 1-year transplantation-related mortality of 20% upon application of reduced-intensity conditioning (RIC). As alloHCT following RIC achieves long-term survival for 50-60% of patients with relapsed or refractory PTCL with otherwise fatal outlook, the benefits of early transplantation outweigh its risks.

The investigations and assessments to be performed throughout the study mainly represent procedures that are performed in PTCL patients according to good medical practice in inpatient and outpatient care. Therefore, these procedures are not expected to pose a specific extended risk to the patient above that of routine care. Most procedures beyond routine patient care such as upfront HLA typing, upfront initiation of a donor search, standardized interim staging by PET/CT, quality-of-life questionnaires and blood-sampling for translational research projects are of added value and do not impose relevant strains.

Moreover, the fact that patients included in this study are closely monitored by an interdisciplinary team, by reference pathologists and reference nuclear medicine physicians and by an independent Data and Safety Monitoring Board, measures that go far beyond routine care, may provide substantial medical benefits by itself.

In summary, there is substantial evidence for a positive benefit-to-risk-ratio for the patients included in this study for both transplant-specific and non-transplant-specific assessments.

Objectives, endpoints and estimands

The primary objective of the REACT study is to assess the progression-free survival (PFS) and the 2-year PFS rate of high-risk nodal PTCL patients. The 2-year PFS rate is an established endpoint in PTCL. The primary endpoint combines chemotherapy-sensitive patients treated by chemotherapy alone and chemotherapy-refractory patients switching from chemotherapy to alloHCT. Chemotherapy refractoriness prompting a protocol-mandated, response-dependent switch to alloHCT is defined as

1. visible lymphoma lesions and SUVmax reduction <50% compared to baseline at iPET, or
2. CT-defined progression on therapy, or
3. a Deauville score of 4 or 5 at EoT PET, or
4. relapse within the first 2 years after the first day of treatment after achieving a complete metabolic response at EoT PET/CT. Based on the literature and the PETAL historical control group, the 2-year PFS rate is estimated to be 40% for patients treated by chemotherapy alone. With up to six cycles of chemotherapy for chemotherapy-sensitive patients and an early switch to alloHCT in chemotherapy-refractory patients, the 2-year PFS rate is expected to rise to at least 55% (Scientific Steering Committee decision). In addition, PFS will be compared to the literature and the PETAL historical control group. Secondary objectives are to assess the remission status at iPET (after 2 cycles of chemotherapy), after the end of chemotherapy and after the end of protocol-based first-line treatment, the response after the end of protocol-based first-line treatment and to assess overall survival, time to progression and disease-free survival. Further secondary objectives are to assess graft-versus-host disease-free and relapse-free survival and the relapse rate. Another secondary objective is to quantify the number of patients allocated to alloHCT, the rate of patients undergoing alloHCT and quality of life. Safety objectives are adverse events and non-relapse mortality. Primary estimand The primary efficacy endpoint is progression-free survival (PFS). The primary estimand is defined accordingly with the following 5 attributes:
5. Population: AlloHCT eligible patients ≥ 18 years diagnosed with previously untreated high-risk nodal peripheral T-cell lymphoma (PTCL) starting treatment, including

• peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS)
• nodal T-follicular helper cell lymphoma (nTFHL)
• anaplastic lymphoma kinase (ALK)-negative anaplastic large cell lymphoma (ALCL)

1. Treatment: Treatment with up to 6 cycles of standard chemotherapy for chemotherapy-sensitive high-risk nodal PTCL patients and response-dependent switch to alloHCT for chemotherapy-refractory high-risk nodal PTCL patients.
2. Variable: PFS status (24 months from treatment start)
3. Intercurrent event strategies:

• Ineligibility ascertained after start of treatment: It will be handled with the treatment policy strategy, including all data regardless of violation of entry criteria.
• Early discontinuation of initially assigned study treatment (not due to progression or death) regardless of the reasons: It will be handled with the treatment policy strategy, including all data regardless of permanent discontinuation of study intervention.
• Additional antilymphoma treatment without prior progression/relapse regardless of the reasons: It will be handled with the treatment policy strategy, including all data regardless of initiation of additional antilymphoma treatment.
• Death of any cause: it will be handled with the composite strategy, counting death as event.

1. Population level summary: PFS probability at 24 months. Study design This is a multicenter, non-randomized, non-blinded clinical study with response-dependent switch to alloHCT for chemotherapy-refractory high-risk nodal PTCL patients. The treatments represent standard procedures for patients with chemotherapy-sensitive disease (chemotherapy) or chemotherapy-refractory disease (immunotherapy), respectively. The study results will be compared with a historical control group (from the phase III PETAL trial) that received first-line chemotherapy without a response-dependent switch to immunotherapy. Standard Treatment PTCL-NOS or nTFHL (including AITL): CHO(E)P regimen ALK-negative ALCL : BV-CHP regimen Allogeneic transplantation Matched related donor (MRD) / matched unrelated donor (MUD): TBI-8/fludarabine regimen or Fludarabine/busulfan regimen (FB8) Haploidentical donor: TBI-8/fludarabine or a protocol based on alkylating agents integrating thiotepa/fludarabine; post-transplantation cyclophosphamide (PTCY) following conditioning therapy is recommended Permitted additional antineoplastic therapies outside the study protocol The above protocols for standard chemotherapy must be adhered to and any deviation must be documented and justified. The alloHCT conditioning regimens represent recommendations. In two instances, additional antineoplastic measures are allowed at the discretion of the investigators:

• Bridging or reinduction therapy before alloHCT
• Consolidation of EoT CMR by high-dose chemotherapy with autoHCT Assessment of quality of life The quality of life of the study participants is evaluated using the EORTC QLQ-C30 Quality-of-Life Questionnaire. Longitudinal assessment of time-to-event endpoints The longitudinal endpoints evaluated in the REACT study rely on PET/CT-defined treatment response criteria (CMR, non-CMR [including partial metabolic response [PMR], no metabolic response [NMR = stable metabolic disease, SMD] and progressive metabolic disease [PMD], CT-defined treatment response categories (CR, PR, OR [CR + PR], SD, PD), relapse after CMR, occurrence and grade of acute or chronic graft-versus-host-disease, death from PD or relapse, non-relapse mortality and death from any cause. Monitoring procedure Monitoring is performed according to ICH-GCP E6 and SOPs to verify that study participants' rights and well-being are protected, reported study data are accurate, complete and verifiable and that the study is conducted in compliance with the currently approved protocol/amendment. Monitoring visits at the study site take place before, during and after the clinical study. Only large cancer centers with departments of hematology, bone marrow transplantation and nuclear medicine [FDG-PET/CT accredited by the European Association of Nuclear Medicine Research Ltd., EANM/EARL] and adequately staffed clinical studies offices will be selected for the study. All study-specific monitoring procedures, monitoring visit frequency and extent of source data verification (SDV) is described in a study-specific monitoring manual. Auditing procedures According to the ICH-GCP guidelines, audits may be performed as a quality measure. Audits may be conducted by the coordinating investigators or an independent external party. Sample size calculation The primary aim of the study regarding efficacy is to estimate the 2-year PFS rate. With up to six cycles of chemotherapy for chemotherapy-sensitive patients and an early switch to alloHCT in chemotherapy-refractory patients, the 2-year PFS rate is hypothesized to be at least 55% (Scientific Steering Committee decision). With a sample size of 70 patients starting treatment within the study, the 95% confidence interval would have a precision of about -/ + 12% and allows excluding a PFS rate at 2 years below 43%, taken into account about 5% patients censored until 2 years. An explorative analysis of patients treated within this study compared to patients from the PETAL trial will be performed. Comparing for example PFS curves between 70 REACT and 50 comparable patients from the PETAL trial (2-year PFS rate about 40%) using the logrank test, an approximately 25% difference (HR=0.47) in 2-year PFS rates can be detected with a power of 80% and a two-sided significance level of 5%. Definition of analysis sets The full-analysis set (FAS) will include all consenting and registered patients receiving at least one treatment component within the study. The per-protocol set (PPS) will include all consenting and registered patients meeting the inclusion criteria, receiving at least one study treatment component and without major protocol violations. The safety analysis set (SAS) will include all consenting and registered patients receiving at least one treatment component within the study and is identical to FAS. Baseline characteristics Demographics and other baseline data will be summarized descriptively. Primary endpoint The primary efficacy analysis will determine the 2-year PFS rate with 95% confidence interval using a Kaplan-Meier curve within FAS. Secondary endpoints The secondary endpoints OS, DFS and GRFS will be analyzed using the same statistical methods as for the primary endpoint. TTP will be analyzed using cumulative incidence curves. Multivariate Cox regression models will be performed for prognostic factors. The rate of remission status based on the PET portion of PET/CT (CMR, non-CMR [including PMR, NMR = SMD, PMD, unknown), the rate of remission status based on the CT portion of PET/CT (CR, PR, OR (CR + PR), SD, PD, or unknown), the rate of response (CR, PR, OR (CR + PR), SD, PD, death or unknown) and the relapse rate will be documented together with the corresponding 95% confidence intervals (Clopper-Pearson). Qualitative endpoints, such as (serious) adverse events, acute and chronic GvHD, secondary malignancies and biological parameters will be presented in frequency tables. Quantitative endpoints will be described in tables displaying sample size, range, mean and standard deviation or median and quartiles, or will be presented graphically. All AEs and SAEs with their severity grade and relationship to treatment will be listed per patient and summarized for all patients together. Subgroup analysis Subgroup analyses regarding sex, IPI risk factors, IPI score, age groups, histological PTCL subtypes, molecular subgroups, allocation to alloHCT or availability of donor are planned. Sensitivity analysis Analyses for efficacy endpoints based on the PPS population are planned as sensitivity analysis in order to assess the robustness of the results obtained from the FAS population. Further sensitivity analyses will be described in the statistical analysis plan (SAP), if necessary. Final analysis The final analysis of the study will be performed after LPLFU and all study documentation is completed as well as the corrections of all CRFs are final (database lock). Biometrical report The results of the final analysis will be presented in a final statistical report. This will provide a description of the patient population, adherence to protocol, the safety of the therapy (particularly in respect of the occurrence of AEs and SAEs), cases of early discontinuation and the results of the efficacy analysis. Scientific Steering Committee (SSC) The Steering Committee of the REACT study takes responsibility for the scientific validity of the study protocol, assessment of study quality and conduct as well as for the scientific quality of the final study report. Data and safety monitoring board (DSMB) The DSMB of the REACT study is composed of four independent experts from the fields of hematological oncology, allogeneic blood stem cell transplantation, nuclear medicine and biostatistics. The DSMB will advise the steering committee to continue, modify or stop the study. Critical non-conformities The term non-conformity refers to the failure to meet specific requirements. It is an umbrella term which includes, for example, protocol deviations, violation of regulatory requirements, non-compliance with internal or external standards, product deficiencies or other undesirable situations. When the study site or any other party involved in the conduct of the study identifies a critical non-conformity, the study site is responsible to inform the coordinating investigators of the clinical study immediately.