Figure Corrections Clarify GSTP1 CAR-T Data in Lymphoma Study
Frontiers in Immunology has issued a correction to the 2025 study on GSTP1 and CAR-T cell function in lymphoma, addressing errors in Figure 2 and Figure 7. The journal states that the mistakes were typographical, data-entry, or figure-compilation problems, and the corrected panels have now replaced the original versions in the article record.
The paper itself centers on a mechanistic question with direct relevance to adoptive cell therapy: whether GSTP1 can improve CAR-T cell proliferation, persistence, and cytotoxicity in a setting of T-cell exhaustion. In that framework, the corrected figures matter because they support the paper’s narrative around oxidative stress, transcriptional control, and antitumor activity in vitro and in vivo. The correction does not introduce new biology. It cleans up the visual and numerical record surrounding a study that linked GSTP1 with impaired exhaustion biology and stronger CAR-T performance against lymphoma.
Figure 2 Repairs the T-Cell Exhaustion Model Readout
The first set of corrections involves Figure 2, which tracks the exhaustion model used to explore GSTP1 expression. In Figure 2A, the word “coasted” was misspelled and has been corrected to “coated.” In Figure 2D, the protein name on the y-axis was corrected from GST1 to GSTP1. Both errors were attributed to spelling mistakes.
Those edits may look minor on the surface, but they sit inside a figure that maps the biological logic of the study. The corrected panel sequence describes CD3+ T cells isolated from healthy donor peripheral blood, activated with anti-human CD3 and anti-human CD28 antibodies, then repeatedly stimulated to generate a T-cell exhaustion model. By day 10, the exhausted phenotype was checked with PD-1 expression, while reactive oxygen species were measured with a DHE probe. GSTP1 expression was then assessed by western blotting, followed by experiments involving shBLIMP1, promoter activity, and chromatin immunoprecipitation.
This part of the dataset matters because exhaustion is not just a descriptive label. Repeated antigenic stimulation pushes T cells toward dysfunction, reduced proliferative capacity, and altered redox balance. GSTP1, a glutathione S-transferase family member, fits naturally into that biology because redox regulation can influence how well T cells survive chronic stimulation and maintain effector function. The corrected figure labeling keeps that chain of reasoning intact.
A single incorrect protein label can redirect interpretation in a pathway map.
Figure 7 Corrects Dose, Imaging Scale, Mouse Placement, and Survival Records
The more consequential correction concerns Figure 7, which presents the in vivo antitumor experiment. Several issues were repaired at once. In Figure 7A, the protocol diagram contained a data-entry error: on day 7, the injected dose of CAR-T cells was originally listed as 7.5×10^7, but the correct dose was 1.5×10^7 CAR-T cells. In Figure 7B, fluorescence scales in the GSTP1 CAR-T group were inconsistent across day 42 and day 52 images during in vivo analysis; the scales were unified across all images. In Figure 7D, mice #3 and #4 in the PBS group and CD19 CAR-T group were repeated because of a figure-compilation error. In Figure 7H, the survival record for mouse #2 in the CD19 CAR-T group on day 82 was incorrect, and the corrected P value between the CD19 CAR-T group and the GSTP1 CAR-T group is 0.2401 rather than 0.1633.
The corrected survival statistic is important, even though it remains nonsignificant. It places the comparison between the CD19 CAR-T group and the GSTP1 CAR-T group on a firmer footing and prevents over-reading a difference that the data do not support. In a cell therapy experiment, dose accuracy and consistent image scaling are not administrative details; they shape how tumor burden, trafficking, and persistence are visualized across groups. When in vivo fluorescence is used to follow disease progression, scale consistency determines whether the reader can judge relative signal strength fairly.
Why the Correction Still Leaves the Biological Claim Intact
The journal states that the errors do not change the scientific conclusions. Based on the correction notice, that is plausible because the underlying issues were not about the study design itself or a reversal of the experimental pattern. The core claim remains that GSTP1 CAR-T cells showed an advantage for antitumor capacity in vivo, while the mechanistic work in the earlier figures connected GSTP1 expression to T-cell exhaustion biology and BLIMP1-dependent transcriptional regulation.
From a clinical immunology standpoint, the study addresses a recurring problem in CAR-T therapy: exhausted cells often lose expansion potential, weaken cytotoxic signaling, and fail to persist long enough to control tumor growth. Any intervention that improves redox balance and limits exhaustion-related dysfunction could, in principle, help a CAR-T product maintain activity inside hostile tumor microenvironments. The correction does not add new evidence for that concept, but it preserves the integrity of the original presentation of the data.
The revised figure set is also a reminder of how much interpretive weight hangs on layout, labeling, and records in translational immunology. A mislabeled axis, an incorrect injected dose, or an inconsistent fluorescence scale can distort the apparent strength of a therapeutic signal, especially when the study depends on imaging, survival curves, and cell-tracking readouts. In a field where small experimental differences can affect conclusions about persistence and potency, precision in figure preparation is part of scientific validity rather than a cosmetic concern.
The corrected article now stands as a cleaner record of the original work, with its mechanistic claims and in vivo comparisons aligned to the updated figures. The remaining question is not whether the correction changes the message—it does not—but how future studies will test whether GSTP1-driven improvements in CAR-T fitness can be reproduced in broader lymphoma models and, eventually, in human therapy settings.