Machine learning predictors for chronic lymphocytic leukemia (CLL) patients obtaining benefit from the addition of obinutuzumab (O) to acalabrutinib (A)

Chronic Lymphocytic Leukemia (CLL) is a type of cancer that affects the blood and bone marrow, commonly seen in older adults. It involves the slow accumulation of white blood cells, leading to impaired immunity and other complications. Traditionally, treatment options include chemotherapy and targeted therapies (cancer treatments that use drugs to precisely identify and attack specific types of cancer cells). Chemotherapy works by using powerful drugs to kill rapidly dividing cells like cancer cells. Targeted therapies work by attacking specific features of cancer cells, like proteins or genes that help them grow, while causing less harm to normal cells.

Recent clinical studies, particularly the ELEVATE-TN trial, have shown that combining the targeted therapies acalabrutinib and obinutuzumab provides significant progression-free survival (PFS - how long a patient lives with the cancer without it getting worse) benefits compared to other treatments. However, while this combination has shown promise, adding obinutuzumab also introduces additional side effects. Prior analyses sought to identify which CLL patient groups benefit most from this combination but did not consider all potential genetic factors.

This study aims to identify which CLL patients benefit the most from combining acalabrutinib with obinutuzumab. By using machine learning—a type of artificial intelligence that finds patterns in large datasets—the study will analyze clinical and genetic information to create a model that predicts treatment response. This will help doctors decide whether adding obinutuzumab is the right choice for individual patients, reducing unnecessary side effects and improving treatment outcomes.

To conduct this research, data from past clinical trials will be analyzed. These trials included patients who received acalabrutinib, either alone or with obinutuzumab. The study will use machine learning methods to examine patient characteristics such as age, blood test results, and genetic markers (specific changes in a patient’s genes, such as IGHV and TP53 mutations, that may affect how their cancer behaves or responds to treatment). By testing the model on different parts of the data, researchers will ensure it provides reliable predictions. The key outcomes measured will be PFS and overall survival (OS - how long people live after being diagnosed with CLL, regardless of the cause of death).

Ultimately, this research seeks to develop an "Optimal Responder Profile" - a way to determine which patients are most likely to benefit from the combination therapy. This will help improve personalized treatment approaches for CLL, leading to better patient care and fewer unnecessary side effects.

慢性淋巴细胞白血病（Chronic Lymphocytic Leukemia, CLL）是一种累及血液与骨髓的恶性肿瘤，多见于老年人群。该疾病以白细胞缓慢增殖蓄积为特征，可引发免疫功能受损及其他并发症。传统治疗方案包括化疗与靶向治疗——靶向治疗指利用药物精准识别并攻击特定癌细胞的癌症治疗手段。化疗通过强效药物杀灭增殖快速的细胞（如癌细胞）；靶向治疗则针对癌细胞的特定特征（如促生长蛋白或基因）发起攻击，对正常细胞的损伤更小。
近期临床研究，尤其是ELEVATE-TN试验结果显示，相较于其他治疗方案，靶向药物阿卡替尼（acalabrutinib）与奥妥珠单抗（obinutuzumab）联合使用可显著延长患者无进展生存期（progression-free survival, PFS——即患者带瘤生存且病情未进展的时长）。尽管该联合疗法展现出良好前景，但加用奥妥珠单抗也会带来额外的不良反应。既往分析尝试明确哪些慢性淋巴细胞白血病患者能从该联合疗法中获益最大，但未纳入所有潜在遗传因素。
本研究旨在明确哪些慢性淋巴细胞白血病患者可从阿卡替尼与奥妥珠单抗的联合治疗中获益最多。本研究将借助机器学习（machine learning，一种可在大规模数据集中挖掘模式的人工智能技术），分析临床与遗传信息，构建预测治疗响应的模型，助力医生判断是否为患者加用奥妥珠单抗，从而减少不必要的不良反应，改善治疗结局。
为开展本研究，将对既往临床试验的数据进行分析。这些试验纳入了接受单药阿卡替尼或阿卡替尼联合奥妥珠单抗治疗的患者。研究将采用机器学习方法，检视患者的年龄、血液检测结果、遗传标志物（即患者基因中可能影响癌症行为或治疗响应的特定改变，如IGHV与TP53突变）等特征。通过在数据集的不同子集上测试模型，研究人员将确保其预测结果具备可靠性。本研究的主要观测终点为无进展生存期与总生存期（overall survival, OS——即慢性淋巴细胞白血病患者自确诊至任意原因死亡的存活时长）。
本研究最终旨在构建"Optimal Responder Profile"——一种可用于判定哪些患者最可能从联合疗法中获益的工具。这将助力优化慢性淋巴细胞白血病的个体化治疗方案，改善患者护理质量，减少不必要的不良反应。