Koos Boeve

40 Chapter 2 Ethical justification As this study used retrospectively evaluated data from patients treated according to the Dutch national guidelines for oral cavity cancer, approval from the hospital research ethics board was not necessary according to the Dutch ethical regulations [8]. Statistics Categorical data are presented as number and percentage, normally distributed data are presented as mean with standard deviation (SD), and skewed data are presented as median with IQR. Fisher’s exact or chi-squared tests were used to test the associations between categorical data. The log-rank test was used to analyse differences between the Kaplan- Meier curves. DSS is reported as a percentage of survival after 5 years and DFS is reported as a percentage of survival after 3 years. The STATA statistical software ( Release 15 . 1) was used to determine the 95% confidence intervals of the DSS and DFS survival percentages (Stata Corp., College Station, TX, USA). All other statistical analyses were performed with ibm spss statistics 23 (SPSS, Chicago, IL, USA). P -values of <0.05 were considered to be significant for all of the statistical analyses. RESULTS Tumour staging In total, 211 patients with a median tumour infiltration depth of 6.0 mm (IQR: 3.3 to 9.9 mm) were used for the pT classification analysis. Tumour restaging according to the 8 th edition resulted in upstaging of 75 (36%) of the 211 patients: 12 (6%) patients on the basis of both tumour infiltration depth and ENE, and 63 (30%) patients on the basis of tumour infiltration depth only. Figure 1A shows the differences between the 7 th and 8 th pT editions. Fifty-four pT1 patients (44%) and 21 pT2 patients (24%) were restaged according to the 8 th edition criteria. Patients with tongue tumours were significantly more often restaged to pT2 (31%) or pT3 (19%) than patients with tumours in the other anatomical locations, for whom restaging occurred in 12% (pT2) and 9% (pT3) (p < 0.001). Of the 38 watchful waiting patients, 11 (29%) were restaged to pT2. These patients had significantly shorter DSS (p = 0.016) and DFS (p = 0.033) than the other 27 patients (Figure 2A, B). Within the watchful waiting group, three (11%) of the 27 non-restaged patients and four (36%) of the 11 restaged patients were diagnosed with regional recurrences during follow-up (not-significant). Sixteen of the 45 patients (35%) restaged from pT1 to pT2 had undergone postoperative radiotherapy after surgical resection of the tumour. Twenty-three of the 30 patients restaged as pT3 had been postoperatively irradiated. The 8 th edition pT classification showed a good stratification with significantly shorter DSS for the pT1-T2 patients upstaged to pT3 than for non-restaged pT2 patients (81% versus 66%, p = 0.048, Figure 3B and Table 2).