ESTRO 2024 - Abstract Book

S1388

Clinical - Head & neck

ESTRO 2024

The investigated primary endpoint was progression free survival (PFS). Uni- and multivariate cox regression analyses were performed for clinical parameters, TLG and (sham)MARD for primary irradiated and surgically treated patients separately. Hazard ratios (HRs) of TLG and MARD were compared using the bootstrap method (10 4 samples) to determine the statistical distribution of (HR1 − HR2), from which the relevant p value then was derived.

Results:

Upon univariate testing several clinical parameters as well as TLG (p<0.001) and the derived MARD (p<0.001) were significantly associated with PFS. Bootstrap comparison of HR between TLG and MARD revealed a significantly better patient stratification by using the best MARD cutoff compared to the best TLG cutoff (p=0.04). Kaplan Meier plots for both parameters are shown in figure 1.

Figure 1: Progression free survival of primary irradiated patients when stratified according to TLG (left) or MARD (right).

Multivariate testing revealed a significant association (p<0.001) of MARD with PFS in primary irradiated patients when accounting for other clinical parameters, results are shown in table 1.

Parameter

Hazard Ratio (confidence interval)

p value

Sex

1.2 (0.925-1.56)

0.17

Age

1.02 (1.01-1.03)

<0.001

T stage

1.4 (1.09-1.8)

<0.001

UICC

1.17 (0.984-1.39)

0.076

MARD

0.996 (0.994-0.998)

<0.001

Table 1: Multivariate testing for progression free survival in irradiated patients.