Table 2. Results of the ROC analysis... | American Society of Hematology

Table 2.

Results of the ROC analysis used to identify optimal cutoff points for PFS and OS in the testing (N = 141) and validation (N = 113) cohorts

	Functional PET parameters	ROC analysis for PFS					ROC analysis for OS
	Functional PET parameters	Cutoff point	Sensitivity, %	Specificity, %	AUC	P	Cutoff point	Sensitivity, %	Specificity, %	AUC	P
Testing set	SUVmax	20	60	55	0.556	.342	24.2	74	40	0.550	.433
	MTV, mL	931	57	75	0.629	.029	1149	52	81	0.670	.011
	TLG	3960	63	62	0.620	.034	6991	57	73	0.661	.012
	MH (AUC-CSH)	0.43	53	66	0.545	.474	0.43	52	69	0.555	.431
Validation set	SUVmax	31	27	86	0.507	.917	31	32	86	0.582	.267
	MTV, mL	336	73	54	0.629	.029	336	79	54	0.637	.070
	TLG	3186	55	66	0.574	.318	3574	63	70	0.634	.079
	MH (AUC-CSH)	0.47	53	66	0.527	.730	0.46	42	85	0.648	.046

	Functional PET parameters	ROC analysis for PFS					ROC analysis for OS
	Functional PET parameters	Cutoff point	Sensitivity, %	Specificity, %	AUC	P	Cutoff point	Sensitivity, %	Specificity, %	AUC	P
Testing set	SUVmax	20	60	55	0.556	.342	24.2	74	40	0.550	.433
	MTV, mL	931	57	75	0.629	.029	1149	52	81	0.670	.011
	TLG	3960	63	62	0.620	.034	6991	57	73	0.661	.012
	MH (AUC-CSH)	0.43	53	66	0.545	.474	0.43	52	69	0.555	.431
Validation set	SUVmax	31	27	86	0.507	.917	31	32	86	0.582	.267
	MTV, mL	336	73	54	0.629	.029	336	79	54	0.637	.070
	TLG	3186	55	66	0.574	.318	3574	63	70	0.634	.079
	MH (AUC-CSH)	0.47	53	66	0.527	.730	0.46	42	85	0.648	.046

Bold P values indicate statistically significant results (P < .05).