Cortellini Alessio, Santo Valentina, Brunetti Leonardo, Garbo Edoardo, Pinato David J, La Cava Giulia, Naidoo Jarushka, Katz Artur, Loza Monica, Neal Joel W, Genova Carlo, Gettinger Scott, Kim So Yeon, Jayakrishnan Ritujith, El Zarif Talal, Russano Marco, Pecci Federica, Di Federico Alessandro, Alessi Joao V, Montrone Michele, Owen Dwight H, Ramella Sara, Signorelli Diego, Fidler Mary Jo, Li Mingjia, Camerini Andrea, Halmos Balazs, Vincenzi Bruno, Metro Giulio, Passiglia Francesco, Yendamuri Sai, Guida Annalisa, Ghidini Michele, D'Alessio Antonio, Banna Giuseppe L, Fulgenzi Claudia A M, Grisanti Salvatore, Grossi Francesco, D'Incecco Armida, Josephides Eleni, Van Hemelrijck Mieke, Russo Alessandro, Gelibter Alain, Spinelli Gianpaolo, Verrico Monica, Tomasik Bartłomiej, Giusti Raffaele, Balachandran Kirsty, Bria Emilio, Sebastian Martin, Rost Maximilian, Forster Martin, Mukherjee Uma, Landi Lorenza, Mazzoni Francesca, Aujayeb Avinash, Dupont Manuel, Curioni-Fontecedro Alessandra, Chiari Rita, Sforza Vincenzo, Tiseo Marcello, Friedlaender Alex, Addeo Alfredo, Zoratto Federica, De Tursi Michele, Cantini Luca, Roca Elisa, Mountzios Giannis, Rocco Danilo, Della Gravara Luigi, Kalvapudi Sukumar, Inno Alessandro, Bironzo Paolo, Di Marco Barros Rafael, O'Reilly David, Fitzpatrick Orla, Karapanagiotou Eleni, Monnet Isabelle, Baena Javier, Macerelli Marianna, Piedra Aida, Agustoni Francesco, Cortinovis Diego Luigi, Tonini Giuseppe, Minuti Gabriele, Bennati Chiara, Mezquita Laura, Gorría Teresa, Servetto Alberto, Beninato Teresa, Lo Russo Giuseppe, Prelaj Arsela, De Giglio Andrea, Rogado Jacobo, Moliner Laura, Nadal Ernest, Biello Federica, Nana Frank Aboubakar, Dingemans Anne-Marie, Aerts Joachim G J V, Ferrara Roberto, Abu Hejleh Taher, Takada Kazuki, Naqash Abdul Rafeh, Garassino Marina Chiara, Peters Solange, Wakelee Heather A, Nassar Amin H, Ricciuti Biagio, Soda Paolo, Caruso Camillo Maria, Guarrasi Valerio
Operative Research Unit of Medical Oncology, Fondazione Policlinico Universitario Campus Bio-Medico, Rome, Italy
Department of Medicine and Surgery, Università Campus Bio-Medico di Roma, Rome, Italy.
J Immunother Cancer. 2025 Sep 29;13(9):e012423. doi: 10.1136/jitc-2025-012423.
With nearly one-third of patients with advanced non-small cell lung cancer (NSCLC) and PD-L1 Tumor Proportion Score≥50% surviving beyond 5 years following first-line pembrolizumab, long-term outcomes challenge traditional paradigms of cancer prognostication. The emergence of non-cancer-related factors and time-dependent trends underscores the need for advanced analytical frameworks to unravel their complex interplay.
We analyzed the Pembro-real 5Y registry, a global real-world dataset of 1050 patients treated across 61 institutions in 14 countries with a long-term follow-up and a large panel of baseline variables. Two complementary approaches were employed: ridge regression, chosen for its ability to address multicollinearity while retaining interpretability, and not another imputation method (NAIM), a transformer-based artificial intelligence model designed to handle missing data without imputation. Endpoints included risk of death at 6, 12, 24, 60 months and 5-year survival.
The ridge regression model achieved a c-statistic of 0.66 (95% CI: 0.59 to 0.72) for the risk of death and an area under the curve (AUC) of 0.72 (95% CI: 0.65 to 0.78) for 5-year survival, identifying Eastern Cooperative Oncology Group Performance Status (ECOG-PS)≥2, increasing age, and metastatic burden as primary risk factors. However, wide CIs for some predictors highlighted statistical instability. NAIM demonstrated robust handling of missing data, with a c-index of 62.98±2.11 for risk of death and an AUC of 60.52±3.71 for 5-year survival. The comprehensive SHapley Additive exPlanations analysis revealed dynamic, time-dependent patterns, with early mortality dominated by acute factors (eg, ECOG-PS, steroids) and long-term outcomes increasingly influenced by systemic health markers (eg, absence of hypertension, increasing body mass index). Unexpected insights included the protective role of dyslipidemia (but not statins) and the nuanced impact of smoking status, reflecting evolving disease dynamics and host-tumor interplay.
Our integrative framework illuminates the complexity of long-term outcomes in patients with NSCLC treated with pembrolizumab, uncovering dynamic, non-linear prognostication trends. This analysis provides insights into patient trajectories, emphasizing the need for holistic, long-term management strategies.
在一线使用帕博利珠单抗治疗后，近三分之一的晚期非小细胞肺癌（NSCLC）患者且程序性死亡受体配体1（PD-L1）肿瘤比例评分≥50%存活超过5年，长期结果挑战了癌症预后的传统模式。非癌症相关因素的出现和时间依赖性趋势凸显了需要先进的分析框架来揭示它们之间复杂的相互作用。
我们分析了Pembro-real 5Y注册研究，这是一个全球真实世界数据集，包含14个国家61个机构治疗的1050例患者，具有长期随访和大量基线变量。采用了两种互补方法：岭回归，因其能够解决多重共线性同时保留可解释性而被选用；以及非另一种插补方法（NAIM），这是一种基于Transformer的人工智能模型，旨在无需插补即可处理缺失数据。终点包括6、12、24、60个月时的死亡风险和5年生存率。
岭回归模型对死亡风险的c统计量为0.66（95%置信区间：0.59至0.72），5年生存率的曲线下面积（AUC）为0.72（95%置信区间：0.65至0.78），确定东部肿瘤协作组体能状态（ECOG-PS）≥2、年龄增加和转移负担为主要风险因素。然而，一些预测因子的宽置信区间突出了统计不稳定性。NAIM展示了对缺失数据的强大处理能力，死亡风险的c指数为62.98±2.11，5年生存率的AUC为60.52±3.71。全面的SHapley加性解释分析揭示了动态的、时间依赖性模式，早期死亡率主要由急性因素（如ECOG-PS、类固醇）主导，长期结果越来越受到全身健康指标（如无高血压、体重指数增加）的影响。意外的发现包括血脂异常（但不是他汀类药物）的保护作用以及吸烟状态的细微影响，反映了疾病动态变化和宿主-肿瘤相互作用。
我们的综合框架阐明了使用帕博利珠单抗治疗的NSCLC患者长期结果的复杂性，揭示了动态的、非线性的预后趋势。该分析为患者病程提供了见解，强调了全面、长期管理策略的必要性。
