The “PET/CT-Guided Early Identification of CAR-T Candidates in Diffuse Large B-Cell Lymphoma: A Multidisciplinary Expert Consensus” integrates opinions from nuclear medicine and hematology, clarifying that [¹⁸F]FDG PET/CT should be upgraded from a purely diagnostic imaging modality to a therapeutic navigation tool, guiding clinicians to identify high-risk patients suitable for chimeric antigen receptor T-cell (CAR-T) therapy as early as possible, thereby avoiding the loss of the optimal treatment window. This article synthesizes the core content of the consensus to help clinicians systematically grasp the key points.
Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma (NHL). Although first-line R-CHOP regimen has significantly improved patient prognosis, relapsed/refractory (R/R) DLBCL has poor outcomes with conventional salvage therapy, and CAR-T therapy offers a curative potential for such patients.
I. Baseline PET/CT: Beyond Staging, Toward Risk Stratification
Compared with conventional contrast-enhanced CT, PET/CT has superior sensitivity and specificity for detecting extranodal involvement, bone marrow infiltration, and lesions with high metabolic activity but without obvious anatomical changes.
l Key quantitative parameters: Metabolic tumor volume (MTV), total lesion glycolysis (TLG), and lesion dissemination index (Dmax) are all significantly associated with poor prognosis in DLBCL patients.
l Reporting and timing requirements: Baseline PET/CT reports should routinely include MTV. The interval between baseline PET and treatment initiation should be as short as possible (recommended within 1 month).
II. Interim PET/CT: The Golden Window for Capturing High-Risk Patients
The consensus recommends that all DLBCL patients undergo interim PET/CT assessment after 3–4 cycles of first-line therapy; if clinical suspicion of disease progression is high, the examination may be performed earlier. The Deauville score (DS) is the cornerstone of interpretation, combined with ΔSUVmax for more precise risk stratification.
1. Deauville Score Stratification Explained
- l DS 1–2: Complete metabolic response (CMR);
- l DS 3: Partial metabolic response (PMR), suggesting favorable prognosis;
- l DS 4–5: Incomplete metabolic response, requiring high vigilance.
The critical watershed in clinical decision-making – DS 4 versus DS 5: DS 5 is one of the indicators suggestive of high-risk disease and treatment failure. DS 4 alone should not trigger treatment escalation or CAR-T referral; it requires integration with other parameters or biopsy confirmation.
2. ΔSUVmax – A Powerful Complementary Quantitative Assessment
ΔSUVmax is defined as the relative percentage decrease in SUVmax from the most active lesion at baseline to the most active residual lesion at interim assessment.
➤ A multicenter prospective study defined:
- l ΔSUVmax ≥66% after 2 cycles or ≥70% after 4 cycles is significantly associated with adequate treatment response and favorable prognosis.
- l When DS 4–5 is combined with ΔSUVmax ≤70%, it identifies an ultra‑high‑risk subgroup with extremely poor prognosis (approximately 10% of patients); these patients should be evaluated for CAR‑T therapy as early as possible.
III. End‑of‑Treatment PET/CT: Confirming Response and Identifying Residual Disease
End‑of‑treatment PET/CT is the standard modality for response assessment after first‑line immunochemotherapy (Class I recommendation, Level A evidence). Its core value lies in distinguishing residual fibrosis from viable tumor, confirming CMR, and early identification of disease progression.
1. DS Remains the Basis of Interpretation, with DS 5 Having Greater Clinical Guidance Value
End‑of‑treatment PET/CT is also interpreted based on DS and the Lugano classification. Consistent with interim assessment, DS 5 is a more reliable parameter for identifying residual disease and poor prognosis.
2. Management Pathway for Patients with Positive End‑of‑Treatment PET (DS 4–5)
① Biopsy of residual lesions should be prioritized to confirm the presence of active disease;
② If biopsy is positive, CAR‑T therapy evaluation should be initiated as early as possible – the consensus defines DS 4–5 after first‑line therapy as a “chemotherapy‑refractory” status, and such patients should be considered potential CAR‑T candidates. If the patient is receiving radiotherapy, radiotherapy may serve as bridging therapy before CAR‑T.
IV. Core Consensus Recommendations: Ten Clinical Action Guidelines
Ten Clinical Action Guidelines
1. PET/CT should serve as a core biomarker in DLBCL management: integrate PET‑derived metrics from diagnosis onward to identify high‑risk patients and facilitate timely CAR‑T referral.
2. Baseline PET/CT should routinely report MTV: using an SUV threshold of 4.0 combined with semi‑automatic segmentation software to measure MTV offers good reproducibility.
3. Interval between baseline PET and treatment initiation should be ≤1 month: DLBCL proliferates rapidly; excessive intervals will affect the true assessment of treatment response.
4. Interim PET/CT should be performed after 3–4 cycles: DS combined with ΔSUVmax can optimize risk stratification and reduce false positives.
5. DS 5 is one of the reliable indicators for early initiation of CAR‑T evaluation: DS 4 should be interpreted with caution, and biopsy confirmation is recommended before treatment escalation.
6. Interim PET reports must include DS, SUVmax, and ΔSUVmax: these metrics are the core basis for clinicians’ stratified decision‑making.
7. Before PET‑suggested change in treatment strategy, biopsy is strongly recommended for confirmation: avoid unnecessary treatment escalation due to false positives.
8. Patients with DS 5 at end‑of‑treatment should be evaluated for CAR‑T as early as possible: DS 4 requires biopsy confirmation before decision‑making.
9. Strictly adhere to standardized PET/CT acquisition protocols: for multiple scans of the same patient, attempt to use the same equipment and reconstruction algorithm to ensure comparability of quantitative parameters.
10. Nuclear medicine physicians and hematologists must collaborate closely: truly integrate PET information into the CAR‑T decision‑making pathway, achieving a paradigm shift from “imaging diagnosis” to “therapeutic navigation”
Conclusion:
Conventional therapy has limited efficacy in R/R DLBCL patients. This multidisciplinary expert consensus systematically defines the CAR‑T‑oriented application of [¹⁸F]FDG PET/CT throughout the entire DLBCL management continuum – from baseline MTV for risk stratification, to interim identification of high‑risk patients, to end‑of‑treatment PET for final confirmation of CAR‑T candidates. PET/CT is evolving from a traditional “diagnostic tool”to a “therapeutic navigator”. The core recommendations of the consensus provide an actionable standardized framework for clinical practice. However, issues such as uniform clinical thresholds for PET quantitative parameters and further validation of the prognostic significance of dynamic changes still require ongoing prospective investigation. It is hoped that deep collaboration between nuclear medicine and hematology will advance this field, enabling more DLBCL patients not to miss this potentially curative opportunity – CAR‑T therapy.
References:
[1] Guerra L, Albano D, Angelillo P, Di Rocco A, Farina M, Farolfi A, Tarantino V, Visco C, Zinzani PL. PET‑guided early identification of CAR‑T candidates in diffuse large B‑cell lymphoma: a multidisciplinary expert consensus. Eur J Nucl Med Mol Imaging. 2026 Jun 26. [Epub ahead of print].