美国国家癌症研究所Grégoire Altan-Bonnet小组的一项最新发现提出了癌症免疫治疗的随机性源于罕见但功能关键的Spark T细胞。2026年2月5日，国际知名学术期刊《细胞》发表了这一成果。

为了评估这些疗法的内在随机性，研究组进行了两组控制良好的体外免疫测定。该研究组表明，白细胞反应和肿瘤细胞毒性在宏观水平上是高度可变的，并且在统计上分布为移位的泊松过程。一种罕见的T细胞亚群（所谓的Spark T细胞）的随机激活，加上旁分泌干扰素(IFN)-γ驱动的正反馈，在免疫治疗反应中解释了这种测量的“噪音”。课题组人员将这些定量见解整合到他们设计的机器学习管道中，以单细胞分辨率分析免疫反应。这导致他们在小鼠幼稚T细胞和人类T细胞母细胞中表型和功能上鉴定了为过继T细胞治疗准备的Spark T细胞。然后，研究人员证明了它们在解释癌症免疫治疗的可变结果方面的相关性。

研究人员表示，癌症免疫疗法在患者和基因相同的小鼠模型中引发高度可变的反应。

附：英文原文

Title: Stochasticity in cancer immunotherapy stems from rare but functionally critical Spark T cells

Author: Emanuel Salazar-Cavazos, Dongya Jia, Yoann Missolo-Koussou, Adam L. Kenet, Sooraj R. Achar, Hannah Dada, Taisuke Kondo, Anagha Krishnan, Naomi Taylor, Nicholas D. Klemen, Peng Jiang, Joshua J. Waterfall, Don L. DeVoe, Grégoire Altan-Bonnet

Issue&Volume: 2026-02-05

Abstract: Cancer immunotherapies trigger highly variable responses in patients and in genetically identical mouse models. To assess the intrinsic stochasticity of these therapies, we performed thousands of well-controlled ex vivo immunoassays. We show that leukocyte responses and tumor cytotoxicity are highly variable at the macroscopic level and statistically distributed as a shifted Poisson process. Stochastic activation of a rare subpopulation of T cells (so-called Spark T cells), coupled with a paracrine interferon (IFN)-γ-driven positive feedback, accounts for this measured “noise” in immunotherapeutic reactions. We integrated these quantitative insights into a custom-designed machine-learning pipeline to analyze immune reactions with single-cell resolution. This led us to phenotypically and functionally identify Spark T cells in murine naive T cells and in human T cell blasts as prepared for adoptive T cell therapy. We then demonstrate their relevance in explaining variable outcomes in cancer immunotherapies.

DOI: 10.1016/j.cell.2025.12.026

Source: https://www.cell.com/cell/abstract/S0092-8674(25)01439-4