Table 3.

Disproportionality approach: sensitivity analyses and relevant signal consistency

	Restricted to the 2015–2020 period	Restricted to non-Japan reports	Corrected for event-related competition bias	Corrected for drug-related competition bias #	Corrected for notoriety bias	Restricted to suspect reports	Signal consistency
Drug	N, ROR (95%CI)	N, ROR (95%CI)	N, ROR (95%CI)	N, ROR (95%CI)	N, ROR (95%CI)	N, ROR (95%CI)
CDK4/6 inhibitors†	161, 1.59 (1.35–1.85)*	83, 1.61 (1.30–1.99)*	122, 1.43 (1.19–1.70)*	154, 1.98 (1.69–2.31)*	122, 1.46 (1.22–1.73)*	158, 1.49 (1.27–1.73)*	STRONG (6/6)
Abemaciclib	60, 9.23 (7.10–11.83)*	11, 3.44 (1.82–5.81)*	45, 9.93 (7.25–13.12)*	60, 11.92 (9.15–15.25)*	42, 9.75 (7.04–13.02)*	58, 8.49 (6.48–10.88)*	STRONG (6/6)
Palbociclib	93, 1.10 (0.90–1.34)	59, 1.38 (1.07–1.77)*	68, 0.94 (0.74–1.18)	87, 1.34 (1.07–1.64)*	72, 1.00 (0.79–1.26)	92, 1.05 (0.84–1.27)	WEAK (2/6)
Ribociclib	14, 1.37 (0.78–2.21)	13, 2.39 (1.34–3.92)*	13, 1.56 (0.88–2.55)	13, 1.70 (0.96–2.77)	13, 1.76 (0.96–2.86)	13, 1.20 (0.68–1.98)	WEAK (1/6)

*Statistically significant disproportionality, i.e., lower limit of the 95% confidence interval > 1 (see text for details). ROR: Reporting Odds Ratio; CI: confidence Interval.

^# Based on the systematic review by Skeoch et al. [6]: amiodarone, methotrexate, leflunomide, cyclophosphamide, bleomycin, gemcitabine, gefitinib, erlotinib, infliximab, etanercept, rituximab, panitumumab, cetuximab, sirolimus, temsirolimus, everolimus, nitrofurantoin, daptomycin

^† The sum of the number of ILD cases for CDK4/6 inhibitors as a drug class is lower than the total number of ILD cases for individual CDK4/6 inhibitors because, in a few reports, more than one agent was recorded