SUPPLEMENTAL DIGITAL CONTENT TABLE 1. Summary of Best Overall Response for FACT-L Total

SUPPLEMENTAL DIGITAL CONTENT TABLE 1. Summary of best overall response for FACT-L total score, TOI and LCS (EFQ Population)

FACT-L, Functional Assessment of Cancer Therapy-Lung; TOI, Trial Outcome index; LCS, Lung Cancer Subscale; EFQ, evaluable-for-quality-of-life; n, number of patients.

a’Improved’ defined as an increase from baseline of ≥6 points for FACT-L and TOI, and ≥2 points for LCS, maintained for ≥21 days.

b’No change’ defined as two visit responses of either ‘no change’, or ‘improved’ and ‘no change’, maintained for ≥21 days, without an intervening visit response of ‘worsened’.

c’Worsened’ defined as a decrease from baseline of ≥6 points for FACT-L and TOI, and ≥2 points for LCS, maintained for ≥21 days.

d‘Other – not evaluable’ defined as a response not qualifying for any of the above.

SUPPLEMENTAL DIGITAL CONTENT FIGURE 1. HRQoL and symptom improvement for FACT-L, TOI and LCS in (A) the overall population, and for (B) patients with EGFR mutation-unknown tumors (EFQ population)

Pre-planned (overall population) and post-hoc (EGFR mutation-unknowns) HRQoL analyses
HRQoL, health-related quality-of-life; FACT-L, Functional Assessment of Cancer Therapy-Lung; EGFR, epidermal growth factor receptor; TOI, Trial Outcome Index; LCS, Lung Cancer Subscale; EFQ, evaluable-for-quality-of-life; CI, confidence interval; OR, odds ratio.
SUPPLEMENTAL DIGITAL CONTENT FIGURE 2. Time-to-worsening in HRQoL and symptoms for (A) FACT-L in the overall population, (B) FACT-L in patients with EGFR mutation-unknown tumors, (C) TOI in the overall population, (D) TOI in patients with EGFR mutation-unknown tumors, (E) LCS in the overall population, and (F) LCS in patients with EGFR mutation-unknown tumors (EFQ population).

Pre-planned (overall population) and post-hoc (EGFR mutation-unknown) calculations. HRQoL, health-related quality-of-life; FACT-L, Functional Assessment of Cancer Therapy-Lung; TOI, Trial Outcome Index; LCS, Lung Cancer Subscale; EFQ, evaluable-for-quality-of-life; G, gefitinib; n, number of patients; C/P, carboplatin/paclitaxel; CI, confidence interval, NC, not calculated.
SUPPLEMENTAL DIGITAL CONTENT FIGURE 3. Mean changefrom baseline in HRQoL and symptoms for (A) FACT-L in the overall population, (B) FACT-L in patients with EGFR mutation-unknown tumors, (C) TOI in the overall population, (D) TOI in patients with EGFR mutation-unknown tumors, (E) LCS in the overall population, and (F) LCS in patients with EGFR mutation-unknown tumors (EFQ population).

Pre-planned (overall population) and post-hoc (EGFR mutation-unknown) calculations. Error bars are the 95% CI of the mean. Data after week 72 not presented for the overall population as <20 patients remain. Data after week 54 not presented for patients with EGFR mutation-unknown tumors as n=2 in the carboplatin/paclitaxel arm of the study with evaluable data at week 60.
FACT-L, Functional Assessment of Cancer Therapy-Lung; TOI, Trial Outcome Index; LCS, Lung Cancer Subscale; EFQ, evaluable-for-quality-of-life; n,number of evaluable patients at baseline; CI, confidence interval.

SUPPLEMENTAL DIGITAL CONTENT FIGURE 4. Kaplan Meier curve for survival without CTC grade 3 or 4 toxicity for (A) the overall population, (B) patients with EGFR mutation-positive tumors, and (C) patients with EGFR mutation-unknown tumors (EFS population).

Pre-planned (overall population) and post-hoc (EGFR mutation-positive and EGFR mutation-unknown) analyses. Time to event calculated from the date of first dose to the earliest date of the following: (i) CTC grade 3 or 4 adverse event, (ii) a worsening from baseline in any laboratory parameter to a CTC grade 3 or 4, or (iii) death.
CTC, Common Terminology Criteria; EFS, evaluable-for-safety; G, gefitinib; C/P, carboplatin/paclitaxel; HR, hazard ratio; CI, confidence interval.