*3.2. Potential Risk Factors of Anti-Cancer Drug-Induced Pneumonia*

The cumulative risk of anti-cancer drug-induced pneumonia was significantly increased in patients with a higher HFS (*p* < 0.0001, Figure 4). The simple Cox proportional hazard analysis showed a risk factor for anti-cancer drug-induced pneumonia (Table 2A); a high HFS and being male were significant risk factors for anti-cancer drug-induced pneumonia. The HFS in particular was a prominent risk factor; an increase of one HFS point was associated with a 16% increased risk of developing anti-cancer drug-induced pneumonia (HR, 1.16; 95% CI, 1.09–1.22; *p* < 0.001). Smoking history and high GS (indicating severe emphysema) were not associated with a risk of anti-cancer drug-induced pneumonia. The robustness of the HFS as a risk factor remained constant in the replaced multiple Cox proportional hazard analyses of other potential risk factors (Table 2B). A multivariate Cox proportional hazard analysis was not performed, because the simple and replaced multiple Cox proportional hazards analyses revealed that the HFS was a prominent risk factor of anti-cancer drug-induced pneumonia.

**Figure 4.** Kaplan–Meier estimated curves of the incidence of anti-cancer drug-induced pneumonia stratified by the high-resolution computed tomography fibrosis score. The Kaplan–Meier estimated curves indicated that the incidence of anti-cancer drug-induced pneumonia is increased in patients with higher high-resolution computed tomography fibrosis scores (HFSs) (*p* < 0.0001 according to the log-rank test).




**Table 2.** *Cont*.

(**A**) † Gene abnormalities includes anaplastic lymphoma kinase (ALK) and epidermal growth factor receptor (EGFR). HR, hazard ratio; CI, confidence interval; ECOG-PS, Eastern Cooperative Oncology Group performance status; n/c, not calculated; GS, Goddard score; HFS, HRCT fibrosis score.


(**B**) HFS, HRCT fibrosis score; HR, hazard ratio; CI, confidence interval; ECOG-PS, Eastern Cooperative Oncology Group performance status; GS, Goddard score.
