Decision analysis for prophylactic cranial irradiation for patients with small-cell lung cancer

Purpose: Prophylactic cranial irradiation (PCI) has been shown to provide survival benefit in patients with limited disease small-cell lung cancer (LD-SCLC) who have achieved complete response. However, PCI may also produce long-term neurotoxicity (NT). The benefits and risks of PCI in LD-SCLC are evaluated. Methods: We developed a decision-analytic model to compare quality-adjusted life expectancy (QALE) in a cohort of SCLC patients who do or do not receive PCI by varying survival rates and the frequency and severity of PCI-related NT. Sensitivity analyses were applied to examine the robustness of the optimal decision. Results: At current published survival rates (26% 5-year survival rate with PCI and 22% without PCI) and a low NT rate, PCI offered a benefit over no PCI (QALE = 4.31 and 3.70 for mild NT severity; QALE = 4.09 and 3.70 for substantial NT severity, respectively). With a moderate NT rate, PCI was still preferred. If the PCI survival rate increased to 40%, PCI outperformed no PCI with a mild NT severity. However, no PCI was preferred over PCI (QALE = 5.72 v 5.47) with substantial NT severity. Two-way sensitivity analyses showed that PCI was preferred for low NT rates, mild NT severity, and low long-term survival rates. Otherwise, no PCI was preferred. Conclusion: The current data suggest PCI offers better QALE than no PCI in LD-SCLC patients who have achieved complete response. As the survival rate for SCLC patients continues to improve, NT rate and NT severity must be controlled to maintain a favorable benefit-risk ratio for recommending PCI.