Background <p>Chemoimmunotherapy outperforms immunotherapy alone in advanced non-small cell lung cancer (NSCLC). However, current anatomical imaging and programmed death-ligand 1 (PD-L1) assessments are insufficient for early response prediction and do not directly capture treatment-induced immune effector function. Whether <sup>68</sup>Ga-grazytracer PET/CT, which visualizes granzyme B as a marker of CD8<sup>+</sup> T‑cell effector function, could assess early response and reveal potential differences between intravenous and bronchial arterial infusion (BAI)-based chemoimmunotherapy remains unknown.</p> Methods <p>In this prospective study, 60 patients with advanced NSCLC receiving BAI-based (<i>n</i> = 28) or intravenous (<i>n</i> = 32) chemoimmunotherapy underwent <sup>68</sup>Ga-grazytracer PET/CT after 1–2 treatment cycles. The primary endpoint was a 6-month objective response (iRECIST). Secondary endpoints compared immune activation between groups and the performance of <sup>68</sup>Ga-grazytracer PET/CT for early response assessment versus conventional biomarkers (baseline tumor size [SLD], early anatomical change [ΔSLD], and PD-L1 expression). Statistical analyses included ROC curve analysis, logistic regression, and bootstrap internal validation.</p> Results <p>BAI-based chemoimmunotherapy showed higher <sup>68</sup>Ga-grazytracer uptake (maximum standardized uptake value [SUV<sub>max</sub>]: 2.86 ± 0.71 vs. 2.19 ± 0.80, <i>P</i> = 0.001) and objective response rate (60.71% vs. 34.38%, <i>P</i> = 0.04) than intravenous chemoimmunotherapy, suggesting a possible association with enhanced immune activation. The SUV<sub>max</sub> demonstrated strong discriminative value for objective response (overall apparent AUC = 0.90; optimism-corrected = 0.89) and was the only independent early-response indicator, outperforming anatomical response metrics (SLD: AUC = 0.57; ΔSLD: AUC = 0.63) and PD-L1 expression (AUC = 0.45). The discriminative accuracy of the SUV<sub>max</sub> was higher in the BAI cohort (apparent AUC = 0.99; corrected = 0.97) than in the intravenous cohort (apparent AUC = 0.83; corrected = 0.83). The corroborative performance of the tumor-to-blood ratio was not significantly different from that of SUV<sub>max</sub>.</p> Conclusions <p><sup>68</sup>Ga-grazytracer PET/CT is a promising early imaging biomarker for the response to chemoimmunotherapy in advanced NSCLC patients, particularly those receiving BAI-based regimens. Larger validation studies with uniform protocols are needed to validate these findings and assess its clinical utility.</p> Trial registration <p>ChiCTR2500105792, the registration date is 2025-07-10, <a href="https://www.chictr.org.cn/showproj.html?proj=275165">https://www.chictr.org.cn/showproj.html?proj=275165</a>.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Improved early response assessment of chemoimmunotherapy efficacy by 68Ga-grazytracer PET/CT in non-small cell lung cancer: a comparison of bronchial arterial versus intravenous chemotherapy

  • Xiuling Shen,
  • Zhengjie Wang,
  • Zhu Xia,
  • Yupin Yi,
  • Wenbo Li,
  • Rui Zhang,
  • Xiaoyang Zhang,
  • Biao Xia,
  • Weibin Jiang,
  • Chunxiao Zheng,
  • Qin Wang,
  • Jia Li,
  • Jianqiu Li,
  • Lu Xu,
  • Zhaofei Liu,
  • Nan Li,
  • Hua Pang

摘要

Background

Chemoimmunotherapy outperforms immunotherapy alone in advanced non-small cell lung cancer (NSCLC). However, current anatomical imaging and programmed death-ligand 1 (PD-L1) assessments are insufficient for early response prediction and do not directly capture treatment-induced immune effector function. Whether 68Ga-grazytracer PET/CT, which visualizes granzyme B as a marker of CD8+ T‑cell effector function, could assess early response and reveal potential differences between intravenous and bronchial arterial infusion (BAI)-based chemoimmunotherapy remains unknown.

Methods

In this prospective study, 60 patients with advanced NSCLC receiving BAI-based (n = 28) or intravenous (n = 32) chemoimmunotherapy underwent 68Ga-grazytracer PET/CT after 1–2 treatment cycles. The primary endpoint was a 6-month objective response (iRECIST). Secondary endpoints compared immune activation between groups and the performance of 68Ga-grazytracer PET/CT for early response assessment versus conventional biomarkers (baseline tumor size [SLD], early anatomical change [ΔSLD], and PD-L1 expression). Statistical analyses included ROC curve analysis, logistic regression, and bootstrap internal validation.

Results

BAI-based chemoimmunotherapy showed higher 68Ga-grazytracer uptake (maximum standardized uptake value [SUVmax]: 2.86 ± 0.71 vs. 2.19 ± 0.80, P = 0.001) and objective response rate (60.71% vs. 34.38%, P = 0.04) than intravenous chemoimmunotherapy, suggesting a possible association with enhanced immune activation. The SUVmax demonstrated strong discriminative value for objective response (overall apparent AUC = 0.90; optimism-corrected = 0.89) and was the only independent early-response indicator, outperforming anatomical response metrics (SLD: AUC = 0.57; ΔSLD: AUC = 0.63) and PD-L1 expression (AUC = 0.45). The discriminative accuracy of the SUVmax was higher in the BAI cohort (apparent AUC = 0.99; corrected = 0.97) than in the intravenous cohort (apparent AUC = 0.83; corrected = 0.83). The corroborative performance of the tumor-to-blood ratio was not significantly different from that of SUVmax.

Conclusions

68Ga-grazytracer PET/CT is a promising early imaging biomarker for the response to chemoimmunotherapy in advanced NSCLC patients, particularly those receiving BAI-based regimens. Larger validation studies with uniform protocols are needed to validate these findings and assess its clinical utility.

Trial registration

ChiCTR2500105792, the registration date is 2025-07-10, https://www.chictr.org.cn/showproj.html?proj=275165.