Table 3. Average number of cycles where paclitaxel-derived grade 1/2 toxicites were registered according to the percentage of short telomeres or average telomere length.

	Average (range) G1/2 episodes
Toxicity type	Patients with critically short telomeres (N = 29)*	Patients without critically short telomeres (N = 86)*	P value (T-test)
Any	4.0 (0–9)	2.2 (0–7)	< 0.001
Myalgia	0.6 (0–3)	0.17 (0–3)	0.005
Peripheral neuropathy	1.2 (0–4)	0.7 (0–3)	0.04
Fatigue	2.2 (0–4)	1.4 (0–4)	0.019
	Patients with average short telomere length (N = 29)***	Patients with average long telomere length (N = 86)****
Any	3.3 (0–9)	2.5 (0–7)	0.100
Myalgia	0.21 (0.3)	0.30 (0–3)	0.55
Peripheral neuropathy	0.57 (0–4)	0.4 (0–3)	0.135
Fatigue	1.9 (0–4)	1.3 (0–4)	0.163

^*Patients with critically short telomeres: patients in whom > 21.9% of their telomeres were shorter than 3 Kb.

^**Patients without critically short telomeres: patients in whom < 21.9% of their telomeres were shorter than 3 Kb.

^***Patients with average short telomeres: in this case, the quartile of patients is selected on the basis of the telomere length distribution. One quartile is still 29 patients (25% of 115 patients); however, the 29 patients defined by the upper quartile of critically short telomeres are not the same patients as those 29 (one quartile) defined by average short telomeres. The cut-off point for average long vs. short telomeres is 8 Kb.

^****Patients with average long telomeres: patients whose average telomere length is > 8 Kb. Similarly to the previous paragraph, these patients are not the same as those 86 patients defined by having few critically short telomeres.