Department of Radiation Oncology, Icahn School of Medicine at Mount Sinai, New York, New York.
Department of Radiation Oncology, Mayo Clinic, Rochester, Minneapolis.
Int J Radiat Oncol Biol Phys. 2023 Jul 15;116(4):858-868. doi: 10.1016/j.ijrobp.2023.01.017. Epub 2023 Jan 21.
Stereotactic radiosurgery (SRS) and immune checkpoint inhibitors (ICI) are highly effective treatments for brain metastases, particularly when these therapies are administered concurrently. However, there are limited data reporting the risk of radiation necrosis (RN) in this setting.
Patients with brain metastases from primary non-small cell lung cancer, renal cell carcinoma, or melanoma treated with SRS and ICI were considered. Time-to-event analyses were conducted for any grade RN and symptomatic RN (SRN) with death incorporated as a competing risk. As a secondary analysis, recursive partitioning analysis (RPA) was used for model development, and a loop of potential models was analyzed, with the highest-fidelity model selected. Brain V12 Gy thresholds identified on RPA were then incorporated into the competing risks analysis. Concurrent SRS and ICI administration.
Six hundred fifty-seven patients with 4182 brain metastases across 11 international institutions were analyzed. The median follow-up for all patients was 13.4 months. The median follow-up was 12.8 months and 14.1 months for the concurrent and nonconcurrent groups, respectively (P = .03). The median patient age was 66 years, and the median Karnofsky Performance Status was 90. In patients with any grade RN, 1- and 2-year rates were 6.4% and 9.9%, respectively. In patients with SRN, 1- and 2-year rates were 4.8% and 7.2%, respectively. On RPA, the highest-fidelity models consistently identified V12 Gy as the dominant variable predictive of RN. Three risk groups were identified by V12 Gy: (1) < 12 cm; (2) 20 cm ≥ V12 Gy ≥ 12 cm; (3) V12 Gy > 20 cm. In patients with any grade RN, 1-year rates were 3.7% (V12 Gy < 12 cm), 10.3% (20 cm ≥ V12 Gy ≥ 12 cm), and 12.6% (V12 Gy > 20 cm); the 2-year rates were 7.5% (V12 Gy < 12 cm), 13.8% (20 cm ≥ V12 Gy ≥ 12 cm), and 15.4% (V12 Gy > 20 cm) (P < 0.001). In patients with any SRN, 1-year rates were 2.4% (V12 Gy < 12 cm), 8.9% (20 cm ≥ V12 Gy ≥ 12 cm), and 10.3% (V12 Gy > 20 cm); the 2-year rates were 4.4% (V12 Gy < 12 cm), 12.4% (20 cm ≥ V12 Gy ≥ 12 cm), and 13.1% (V12 Gy > 20 cm; P < 0.001). There were no statistically significant differences in rates of any grade RN or SRN when accounting for therapy timing for all patients and by V12 risk group identified on RPA.
The use of SRS and ICI results in a low risk of any grade RN and SRN. This risk is not increased with concurrent administration. Therefore, ICI can safely be administered within 4-weeks of SRS. Three risk groups based on V12 Gy were identified, which clinicians may consider to further reduce rates of RN.
立体定向放射外科（SRS）和免疫检查点抑制剂（ICI）是治疗脑转移瘤的有效方法，特别是当这些治疗方法同时使用时。然而，目前关于这种情况下放射性坏死（RN）风险的数据有限。
本研究纳入了接受 SRS 和 ICI 治疗的原发性非小细胞肺癌、肾细胞癌或黑色素瘤脑转移患者。采用时间相关事件分析方法，分析任何级别 RN 和症状性 RN（SRN）的发生情况，并将死亡作为竞争风险纳入分析。作为次要分析，采用递归分区分析（RPA）进行模型开发，并对潜在模型的循环进行分析，选择最高保真度的模型。然后将 RPA 确定的脑 V12Gy 阈值纳入竞争风险分析中。SRS 和 ICI 同步治疗。
本研究共纳入了 11 个国际机构的 657 名患者的 4182 个脑转移瘤。所有患者的中位随访时间为 13.4 个月。同步组和非同步组的中位随访时间分别为 12.8 个月和 14.1 个月（P=0.03）。中位患者年龄为 66 岁，Karnofsky 表现状态评分为 90 分。在任何级别 RN 患者中，1 年和 2 年的发生率分别为 6.4%和 9.9%。在 SRN 患者中，1 年和 2 年的发生率分别为 4.8%和 7.2%。在 RPA 中，最高保真度的模型一致地将 V12Gy 确定为预测 RN 的主要变量。根据 V12Gy，可将患者分为 3 个风险组：（1）<12cm；（2）20cm≥V12Gy≥12cm；（3）V12Gy>20cm。在任何级别 RN 患者中，1 年的发生率分别为 3.7%（V12Gy<12cm）、10.3%（20cm≥V12Gy≥12cm）和 12.6%（V12Gy>20cm）；2 年的发生率分别为 7.5%（V12Gy<12cm）、13.8%（20cm≥V12Gy≥12cm）和 15.4%（V12Gy>20cm）（P<0.001）。在任何级别 SRN 患者中，1 年的发生率分别为 2.4%（V12Gy<12cm）、8.9%（20cm≥V12Gy≥12cm）和 10.3%（V12Gy>20cm）；2 年的发生率分别为 4.4%（V12Gy<12cm）、12.4%（20cm≥V12Gy≥12cm）和 13.1%（V12Gy>20cm；P<0.001）。对于所有患者和根据 RPA 确定的 V12 风险组，无论治疗时机如何，SRS 和 ICI 同时使用并不会增加任何级别 RN 或 SRN 的发生率。
SRS 和 ICI 的联合应用导致任何级别 RN 和 SRN 的风险较低。这种风险不会因同时使用而增加。因此，ICI 可以在 SRS 后 4 周内安全使用。根据 V12Gy 确定了 3 个风险组，临床医生可以考虑进一步降低 RN 发生率。
