Outcomes of Patients Intended for Autologous Stem Cell Transplantation with CT-Based Partial Response but PET-Based Deauville Score of 5 after Salvage Chemotherapy
Bibliographic record
Abstract
Background: Salvage chemotherapy (ST) followed by autologous stem cell transplant (ASCT) is widely used to treat relapsed/refractory large B-cell lymphomas (LBCL), both for late relapses and in jurisdictions without access to second-line chimeric antigen receptor T-cell (CAR-T) therapy. The most important criterion in determining ASCT eligibility is response to ST defined historically as CT-based complete or partial morphologic response (CR, PR) and more recently, complete or partial metabolic response (CMR, PMR) on pre-ASCT PET according to the Deauville Score (DS). Pre-ASCT DS has been shown to predict outcomes, but this has mainly been shown in patients (pts) with DS1-3 or 4. For pts with a PR but DS5, it is unknown whether the most appropriate pathway is to pursue ASCT or to move to third-line CAR-T. Our objective was to assess the outcomes of pts with PR with DS5 on pre-ASCT PET, comparing those who received ASCT to those who moved to third-line CAR-T, against a control group of pts with PR with DS4. We also explored whether delta (Δ) SUVmax >66%, a validated interim response measure in frontline LBCL, performed better than DS in discriminating outcomes. Methods: Retrospective study of pts intended for ASCT at Princess Margaret Cancer Centre from January 2014 to July 2024. Inclusion criteria: adults >18 years with a histological diagnosis of de novo or transformed LBCL receiving ST (anthracycline or platinum-based) intended for ASCT with a pre-ASCT PET/CT (typically performed after 3 cycles of (R)GDP or 4-6 cycles of anthracycline-based treatment for transformed disease) demonstrating PR (per RECIST criteria) with DS5 (SUVmax >3 times liver). A control group of pts with PR with DS4 was included. Outcomes included progression-free survival (PFS) (date of pre-ASCT PET to progression or death after cell infusion (stem cells or CAR-T cells)) and overall survival (OS) from pre-ASCT PET. Survival probability and effect sizes were calculated using Kaplan-Meier method and cox regression. A two-sided p-value of 0.05 was used to establish significance. Results: We included 105 pts with PR on pre-ASCT PET (45: DS5, 60: DS4). Of the 45 pts with DS5, 36 proceeded to ASCT and 9 to CAR T. There were no differences in baseline characteristics (age, sex, diagnosis, refractory status, revised international prognostic index (rIPI), stage, bulk and persistently FDG avid sites) between the 3 groups (ASCT DS4, ASCT DS5, CAR-T DS5). Median age was 58 yrs (IQR 46-63); 34% were female. De novo diffuse large B-cell lymphoma (DLBCL), high grade B-cell lymphoma, transformed DLBCL and primary mediastinal B-cell lymphoma comprised 54%, 9%, 30% and 7% of cases, respectively. Primary refractory disease (<3 months), early relapse (3-12 months), late relapse (>12 months), and POD24 with transformed DLBCL were seen in 56%, 12%, 28%, and 4% of cases, respectively. At time of relapse, 70% had stage III/IV, 39% had bulky disease (>10 cm) and 32% had high risk rIPI. Pre-ASCT PET SUVmax was similar between the ASCT DS5 and CAR-T DS5 groups; 14.9 (95% CI 12-19.5) vs. 19.4 (95% CI 17.1-24.2), p=0.26. All pts who underwent CAR-T received axicabtagene ciloleucel: 6/9 received radiation bridging, 2/9 received steroid bridging and 1/9 received no bridging. During a median follow up of 424 days (IQR 214-1249), 56 (53%) pts progressed and 44 (42%) died. The 6-month PFS for pts receiving ASCT DS4, ASCT DS5, and CAR-T DS5 was 73% (95% CI 60-84), 46% (95% CI 30-63), and 63% (95% CI 29-96), p<0.0001. Among pts with DS5, CAR-T vs. ASCT demonstrated a HR of 0.77 (95% CI 0.32-1.89), p=0.56. ΔSUVmax >66% did not discriminate outcomes better than DS in pts receiving ASCT. Of pts who progressed after ASCT when CAR-T was available, 63% (17/27) received third-line CAR-T. There was no difference in 6-month OS between the three groups; ASCT DS4 93% (95% CI 86-100), ASCT DS5 77% (95% CI 63-91) and CAR-T DS5 88% (95% CI 65-100), p=0.08. Conclusion: Our data represent a sizable cohort of pre-ASCT PET DS5 pts. Outcomes in the CART D5 group were not significantly different compared to the ASCT DS group, which may be related to inadequate statistical power with small sample size. Poor outcomes are noted in pts with pre-ASCT PET DS5 disease regardless of therapeutic approach representing an area in need of improved treatment. The analysis will be updated and we intend multicentre collaboration, as determining potential differences between D5 groups could impact clinical practice.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".