Defining lines of therapy in haematological malignancies: a proposed systematic and comprehensive framework from the EBMT Practice Harmonisation and Guidelines Committee

Abstract

The therapeutic landscape of hematological malignancies has been transformed by targeted therapies, immunotherapy, and cellular therapies, leading to more complex and prolonged clinical courses. Currently, there is substantial variability in how “lines of therapy” (LoT) are defined and recorded, which hinders research and data harmonization. To address this, the EBMT Practice Harmonisation and Guidelines Committee has developed a standardized framework for LoT across acute leukemias, lymphomas, and multiple myeloma. A LoT is defined as a coherent therapeutic episode aimed at achieving or maintaining disease control, comprising one or more systemic treatment phases such as induction, consolidation, or maintenance, administered in the absence of disease progression, treatment failure, or major toxicity. A new LoT is assigned at documented progression, relapse, or primary refractory disease, and the introduction of an agent from a distinct therapeutic class to deepen response also constitutes a new line, whereas dose adjustments, schedule changes, or planned omissions to manage toxicity do not create a new LoT. Bridging therapies administered prior to cellular therapy, including CAR-T or transplantation, are considered part of the same LoT as the definitive treatment. This initiative aims to enhance methodological rigor, improve clarity in treatment reporting, and support evidence-based decision-making in precision hematology.

Questions asked/issues to be addressed

What defines a Line of Therapy (LoT) in hematological malignancies, and why is a standardized definition crucial for clinical practice and research?

How has the emergence of targeted, immune, and cellular therapies redefined traditional approaches to LoT?

How does variability in LoT classification across studies, registries, and clinical practice impact patient care and research?

Which principles should be used to define a LoT?

What are the most recent published guidelines for the different hematological malignancies?

Current state of the art

The therapeutic management of hematological malignancies comprises a series of interdependent, sequenced treatment phases. Recent advances in molecularly targeted agents and immune-based therapies have profoundly altered the clinical trajectory and natural history of many hematologic neoplasms. Diseases once characterized by rapid progression and dismal outcomes now exhibit substantially improved survival, frequently extending over multiple sequential lines of therapy (LoT) administered across prolonged disease courses.

The current definition in the EBMT registry states that a line of treatment covers all cycles of the same drugs or regimen administered during the same period, for the same reason, and within the same disease status of the patient. Within this increasingly complex therapeutic landscape, EBMT has identified the need to further standardize the definition of LoT across hematologic malignancies. By formulating consensus-based criteria for LoT determination, the EBMT aims to harmonize data collection, improve the clarity of treatment reporting, and support the integration of emerging therapies into routine practice. Substantial variation currently exists in how LoTs are defined and counted across studies and registries. In the absence of standardized guidelines, clinicians and investigators often rely on center-specific and individual clinician/ data manager interpretations, leading to marked variability and potential misclassification, thereby potentially hindering routine clinical care, registry-based research, and clinical trial eligibility. Establishing uniformity is therefore a critical step towards enhancing methodological rigor, strengthening registry harmonization, and advancing evidence-based decision-making in precision hematology.

The challenges inherent in defining and enumerating LoTs are particularly well illustrated by the experience in multiple myeloma, a disease characterized by prolonged survival and inevitable exposure to multiple sequential therapies. In 2015, Rajkumar and colleagues proposed guidelines for determining the number of prior LoT in multiple myeloma [1].

Although these myeloma-specific guidelines represented a major step toward standardization and have since been widely adopted in clinical trials and regulatory contexts, they also highlight broader challenges that extend across the breadth of hematologic malignancies. As therapeutic strategies increasingly incorporate complex combinations, multiple sequential phases of treatment, and novel modalities including cellular and immune-based therapies, definitions based solely on regimen changes or treatment discontinuation may fail to capture therapeutic intent, the biological significance of disease control or failure, and the continuity of planned treatment strategies.

In this article, we present our contemporary perspective on a systematic and comprehensive framework for determining LoT in a consistent manner across hematologic malignancies in 2026 and beyond.

Methodology

This framework was developed following the EBMT Practice Harmonisation and Guidelines Committee methodology [2]. A panel of hematology and cell therapy experts from the EBMT, including the chairs of the Cellular Therapy and Immunobiology Working Party (CTIWP), Lymphoma Working Party (LWP), Pediatric Diseases Working Party (PDWP), Acute Leukemia Working Party (ALWP), and Chronic Malignancies Working Party (CMWP), was convened to address the need for standardized definitions of LoT across hematological malignancies (acute leukemia, lymphomas, and multiple myeloma). The scope of hematological diseases and disease-specific recommendations was discussed during three dedicated teleconferences in autumn 2025 and consolidated in a one-day, in-person workshop held in October 2025.

Recommendations were iteratively developed based on expert consensus, clinical experience, literature review, and analysis of EBMT registry data. Disease-specific examples were generated for acute leukemia, lymphomas, and multiple myeloma. Drafts were circulated electronically and refined through multiple rounds of review by all coauthors.

Workshop recommendations

Definition of line of therapy

We begin by defining key terms required to establish a LoT accurately.

A LoT is a coherent therapeutic episode initiated to achieve, re-achieve, or maintain disease control, characterized by a dominant intent (curative, disease-controlling, or palliative), and composed of one or more sequential systemic treatment phases administered without occurrence of disease progression or treatment failure or major toxicity requiring a change in treatment.

This definition is anchored in two conceptual pillars:

1. Therapeutic Intent: the clinical objective guiding treatment initiation determines the categorization of a LoT.

2. Disease progression: sequential treatment phases administered in the absence of disease progression or treatment failure constitute a single LoT. The initiation of new therapy following clinically confirmed progression defines the start of a subsequent LoT.

Systemic treatment is commonly structured into distinct but sequentially connected phases—induction, consolidation, maintenance, bridging, and salvage therapy—each defined by its intent and temporal relationship to disease response or progression.

Induction constitutes the initial therapeutic phase, aimed at achieving remission through the eradication or substantial reduction of malignant cells. Once remission is achieved, consolidation therapy follows to eliminate residual disease and deepen response, often employing intensified chemotherapy, targeted agents, and haematopoietic cell transplantation (HCT) or cellular therapies. Maintenance therapy is subsequently administered to sustain remission and delay relapse, typically through lower-intensity or continuous regimens designed to suppress residual clonal activity.

Other therapeutic treatment phases include holding or bridging therapy that encompasses any systemic treatment delivered before a planned definitive therapeutic intervention, such as chimeric antigen receptor – T cells (CAR-T). Holding therapies are defined as those administered between the identification of the need for CAR-T and leukapheresis, and bridging therapies are those delivered between leukapheresis and subsequent CAR-T infusion. These treatments are intended to maintain or achieve a remission, prevent disease regrowth, or stabilize the patient during logistical or medical delays that precede the definitive therapy of CAR-T cell collection, manufacturing, and reinfusion.

In contrast, salvage therapy refers to the administration of a new regimen following relapse, progression, or treatment failure, with the objective of reinducing remission, prolonging disease control, or providing palliation. A LoT is considered ended when there is disease progression, relapse, or treatment failure. Treatment failure due to persistence of disease is defined as the continued presence of detectable disease following a planned therapy, indicating that the treatment did not achieve adequate disease control or eradication. In such cases, the new treatment initiated is defined as salvage therapy and marks the beginning of a new LoT.

Therefore, the total number of therapeutic lines prior to cellular therapy (allogeneic (allo)-HCT, autologous (auto)-HCT, or CAR-T) is calculated as: First-line treatment + number of salvage therapies received.

EBMT criteria for LoT assignment

Clarifications and disease-specific examples (acute leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, and multiple myeloma)

1. Induction, consolidation, and maintenance therapies are considered to be components of a single, continuous LoT when delivered without intervening progression and with a consistent therapeutic intent.

   Clarification: all planned treatment phases constituting parts of a unified strategy remain within the same LoT, provided no measurable residual disease (MRD)-defined or overt progression occurs.

   Example: In AML, induction chemotherapy followed by consolidation, up to four cycles (including allo-HCT/auto-HCT), and subsequent maintenance strategy is considered a single LoT (Fig. 1a) [3].

2. A new LoT must be assigned when progression, relapse, or primary refractory disease is documented.

   Clarification: Loss of disease control—whether detected clinically, morphologically, radiologically, or by validated MRD assays—indicates the end of the preceding therapeutic episode. This includes not only overt progression or relapse after response, but also failure to achieve disease control at the end of treatment, defined as primary refractory disease. In all such cases, initiation of a new therapeutic strategy constitutes a new LoT. (Example AML- Fig. 1b, example ALL - Fig. 1c) [3].

   In the case of Diffuse Large B-Cell Lymphoma (DLBCL), persistent disease at the end of treatment is termed primary refractory disease. CAR-T therapy or salvage therapy, e.g., R-DHAP and auto-HCT, is considered second-line therapy despite the absence of progression (Fig. 2a, b) [4].

   In multiple myeloma, a patient who progresses on post-transplant maintenance Daratumumab and Lenalidomide and whose treatment is switched to a Carfilzomib-based regimen is starting on a second line of therapy (Fig. 3a) [5].

3. The addition of therapeutic agents that are pre-planned and incorporated a priori into the initial treatment regimen does not constitute a new LoT, even when these agents belong to different pharmacologic classes.

   Clarification: Some contemporary regimens incorporate targeted or immunotherapeutic agents as part of a predefined treatment strategy. Although these agents may have different mechanisms of action, they remain part of a single therapeutic episode when they are prospectively pre-planned and specified in the initial treatment protocol. If the treatment plan is prospectively defined, initiated before evidence of disease progression, and delivered within the same intent to achieve remission, the entire regimen constitutes one single LoT, irrespective of the number or diversity of agents included.

   Examples: In AML, gemtuzumab ozogamicin, venetoclax, or other targeted therapies may be added upfront to induction and consolidation chemotherapy as part of the standard induction regimen. These agents are all considered part of the first-line therapy and therefore do not constitute a new LoT [6, 7].

   In frontline B- acute lymphoblastic leukemia (ALL) treatment, blinatumomab given early as a planned MRD-directed consolidation step remains part of the initial treatment strategy and does not represent a new LoT [8].

   An insufficient response leading to a planned treatment escalation is considered part of the same LoT, provided there is no disease progression. For example, in Hodgkin lymphoma, a patient treated with ABVD who does not achieve a metabolic complete response on interim PET, escalation to BEACOPP or BrECADD is part of the same LoT. Similarly, radiotherapy for residual disease is considered part of the same LoT when it is pre-planned (Fig. 2c) [9].

4. The introduction of an agent from a different therapeutic class—added with the purpose of improving or deepening response—initiates a new LoT.

   Clarification: the addition of a pharmacologically distinct agent reflects a substantive modification of the therapeutic strategy. When such an agent is introduced with the explicit aim of augmenting response, converting partial response to complete remission, eliminating MRD, or achieving deeper disease control, the treatment plan transitions to a new LoT.

   Example: In ALL, a patient treated with multi-agent chemotherapy who achieves morphological remission but remains MRD-positive and subsequently receives blinatumomab with the intent of eradicating MRD and deepening response is considered to have initiated a new LoT. The introduction of a bispecific T-cell engager, by way of example, represents a change in therapeutic class and reflects a new therapeutic strategy rather than continuation of the initial regimen [10].

5. The modification of dose, schedule, or timing does not change the LoT.

   Clarification: Adjustments to dosage, administration schedule, or treatment timing—including delays, interruptions, or spacing of cycles—whether performed for toxicity management, organ dysfunction, intolerance, patient frailty, or intercurrent illness, do not constitute a new LoT. Similarly, partial omission or early discontinuation of specific components of a regimen remains within the same LoT, provided that the overall therapeutic intent and fundamental regimen structure are preserved. Such modifications are recognized components of routine clinical practice aimed at optimizing safety, tolerability, and feasibility without altering the underlying therapeutic strategy. Accordingly, treatment continues to be classified under the same LoT.

   Example: In Hodgkin lymphoma, a patient treated with ABVD in whom an interim PET scan after two cycles (PET2) demonstrates complete metabolic response may have bleomycin omitted from subsequent cycles (i.e., continuation with AVD). This planned omission to reduce toxicity while preserving the original therapeutic intent does not constitute a new LoT and remains part of the same LoT (Fig. 2d) [11].

6. The substitution of an agent with an agent of a different spectrum constitutes a new LoT.

   Clarification: If a therapeutic agent is discontinued and replaced by another agent with a different pharmacologic spectrum or mechanism of action, this change represents a substantive modification of the therapeutic strategy and therefore defines the initiation of a new LoT. This applies irrespective of whether the substitution is prompted by toxicity, intolerance, or lack of efficacy, provided that the change reflects a shift in the dominant treatment approach rather than a modification of dose or schedule within the same regimen.

   Examples: A patient with Philadelphia chromosome (Ph) + ALL receiving dasatinib develops a clinically significant pleural effusion, necessitating discontinuation of the drug. Therapy is subsequently switched to ponatinib, a third-generation tyrosine kinase inhibitor (TKI) with a broader inhibitory spectrum and activity against resistant disease-associated ABL mutations. This substitution constitutes a new LoT (Fig. 1d) [12].

   In the case of Multiple myeloma, if a patient develops severe hepatitis on an anti-B-Cell Maturation Antigen (BCMA)bispecific antibody and then switches to an alternative anti-G protein-coupled receptor class C group 5 member D (GPRC5D) bispecific antibody, this constitutes a new LoT (Fig. 3b) [13].

7. Holding and Bridging Therapy Before Cellular Therapy Remains Within the Same LoT.

Fig. 1. Disease-specific examples for acute leukemias.

a AML example. One LoT., b AML example. Two LoT in AML. c ALL example two LoT. d Example Ph+ ALL - TKI toxicity. Two LoT.

Fig. 2. Disease-specific examples for lymphomas.

a DLBCL example two LoT. b Example DLBCL CAR-T as second LoT. c Hodgkin Lymphoma examples one LoT. d Hodgkin Lymphoma example one LoT.

Fig. 3. Disease-specific examples for multiple myeloma.

a Multiple Myeloma example two LoT. b Example Multiple Myeloma – treatment toxicity. Two LoT. c Example Multiple Myeloma – Holding and Bridging therapy before CAR-T – One Line.

Clarification: Bridging therapy administered prior to lymphodepletion and infusion of cellular therapy (e.g., CAR-T cells) does not constitute a new therapeutic episode, as it does not alter the dominant therapeutic intent, which remains focused on proceeding to definitive cellular therapy. Bridging treatment is therefore considered a functional component of the same LoT during short intervals required for CAR-T manufacturing, recovery from toxicity, completion of eligibility assessments, or optimization of organ function. In this framework, the entire sequence of eventual holding and bridging therapy followed by CAR-T infusion constitutes a single LoT, even though multiple agents or regimens may be used within that interval. The LoT is defined by the overarching therapeutic strategy (delivery of cellular therapy), not by the number of interim regimens administered. While bridging therapy and CAR-T together comprise a single LoT, the line number assigned to CAR-T reflects the number of prior completed LoTs. Thus, CAR-T may be described as being administered after two prior lines of therapy, while still constituting the current LoT together with any associated bridging therapy.

Examples: In diffuse large B-cell lymphoma (DLBCL), a patient who experiences early relapse after first-line R-CHOP is planned for CAR-T cell therapy. The patient receives bridging chemotherapy to control the disease while awaiting CAR-T manufacturing, followed by lymphodepletion and CAR-T infusion. In this scenario, both the bridging therapy and the CAR-T cell infusion are considered part of the same second LoT, as they are components of a single therapeutic strategy. At the same time, CAR-T is correctly described as being administered after one prior LoT, consistent with regulatory and clinical trial conventions (Fig. 2b) [4, 14].

A patient with multiple myeloma experiences relapse and is started on Carfilzomib, Daratumumab, and Dexamethasone (KDd) as the next line of therapy. Following lymphocyte collection, the patient receives Talquetamab as bridging therapy prior to conditioning and a Ciltacabtagene Autoleucel (Cilta-cel) infusion. The Talquetamab does not constitute a new LoT (Fig. 3c) [15].

Upcoming research and harmonization recommendations

Additional harmonization efforts are warranted to validate this framework across a broader range of hematological diseases. We believe that the framework proposed in this manuscript could represent a useful step toward future harmonization of LOT reporting across transplant and cell therapies registries. Prospective evaluation within the EBMT registry will support assessment of data quality, consistency, standardization, and real-world applicability. With the aim of further improving the quality of the EBMT registry, a fully updated data collection manual will be distributed to centers, reflecting the revised definitions and procedures for reporting treatment lines, and educational webinars for clinicians and data managers will be delivered.

Author contributions

AR, AB: leaders of the workshop. AR, AB, FO, ISO, and IY-A conceived and designed the manuscript topics to be covered; AC drafted the first version of the manuscript. JEG provided the statistical and technical inputs. All coauthors played an important role in drafting the recommendations and preparation of the figures, revising the manuscript, approving the final version, and agreeing to be accountable for all aspects of the work. No generative AI tools were used in the preparation of this manuscript.

Funding

The EBMT is a not-for-profit medical and scientific organization. The EBMT performed administrative and legal management, as well as funding acquisition. No pharmaceutical company and no other funding source were involved in any way or had a role in the study design, data collection, data analysis, data interpretation, or writing of the report. No medical writer or editor was involved.

Competing interests

The authors declare no competing interests.