paediatrics Brussels 17

298

Kennedy et al.

International Journal of Radiation Oncology Biology Physics

Table 3 Outcome z -scores by treatment group in all participants stratified by age into those aged < 8.0 years and those aged 8.0 years at diagnosis

< 8.0 y at diagnosis

8.0 y at diagnosis

Intergroup mean difference (95% CI)

Intergroup mean difference (95% CI)

Outcome

n1, n2 HFRT

STRT

P

n1, n2 HFRT

STRT

P

Executive function (BRIEF)

29, 24 0.45 (0.83) 0.39 (1.11) 0.84 (0.31 to 1.38)

.003 39, 47 0.09 (0.88) 0.16 (1.03) 0.25 ( 0.17 to 0.67) .24

Health status (HUI3)

26, 21 0.14 (0.76) 0.09 (1.13) 0.23 ( 0.32 to 0.79) .41 29, 38 0.10 (0.94) 0.17 (1.22) 0.27 ( 0.28 to 0.82) .33

Behavioral

31, 23 0.18 (0.75) 0.30 (1.17) 0.48 ( 0.09 to 1.04) .10 19, 27 0.02 (1.09) 0.05 (1.03) 0.03 ( 0.61 to 0.67) .93

difficulties (SDQ)

Quality of life

20, 17 0.18 (1.04) 0.06 (1.01) 0.23 ( 0.45 to 0.92) .50 42, 49 0.02 (1.01) 0.07 (0.98) 0.10 ( 0.32 to 0.51) .64

(PedsQL and QLQ-C30)

Height decrement from diagnosis Weight decrement from diagnosis

23, 17 1.62 (0.85) 0.91 (0.84) 0.71 ( 1.26 to 0.17) .012 36, 39 1.05 (0.88) 0.82 (0.89) 0.23 ( 0.64 to 0.18) .26

24, 19 0.23 (1.13) 0.02 (1.04) 0.25 ( 0.92 to 0.43) .47 35, 41 0.53 (0.93) 0.31 (0.84) 0.21 ( 0.62 to 0.19) .30

Abbreviations: PedsQL Z Quality of Life Inventory; QLQ-C30 Z core 30-item version of the Quality of Life Questionnaire. Other abbreviations as in Tables 1 and 2 . Values are mean (SD).

tendency for epiphyseal tissue to respond to HFRT as an “early reacting” tissue, or both. An interaction between the effect of

HFRT and its use in combination with chemotherapy cannot be ruled out. The estimated biologically equivalent dose (BED) of HFRT relative to STRT on the CNS was calculated on the assumption, based on rates of radiation-induced necrosis within the CNS, that a / b Z 2 for the linear ( a ) and quadratic ( b ) components of in- crease in the radiation dose per fraction on the CNS in the linear- quadratic radiobiological model (26) . On this calculation, the BED at 2 Gy of HFRT for the CNS was 21.4% higher for the craniospinal dose outside the boost field (27.0 vs 22.2 Gy BED), unchanged for the tumor bed (51.0 vs 51.8 Gy BED), and 15% lower in the remaining posterior fossa boost field, which included areas of the cerebral cortex adjacent to the cerebellum (45.1 vs 51.8 Gy). The effect on CNS outcomes of these interarm differ- ences is difficult to predict. The better executive function reported here is consistent with the encouraging cognitive function reported in previous uncontrolled studies (27, 28) , but the absence of dif- ferences in health status, behavior, and QoL suggests similarity of treatment arms with respect to effects on the CNS. Neither the absence of difference between treatments in PNET4 with respect to Event free survival (EFS) (30) nor the greater decrement in height after HFRT reported here were predicted in advance of PNET4. In conclusion, this study highlights the uncertainty of radio- biological assumptions with respect to early- and late-reacting tissue components in the normal CNS and bony spine. Although the present study suggests some benefit to executive function associated with HFRT, it also showed an absence of associated benefit to behavior, health status, or quality of life and does not enable us to reach a final conclusion on whether HFRT was of greater overall benefit than STRT to QoS. Collation of neuro- psychometric testing, collected within some participating national groups on PNET4 survivors, into an international dataset is in progress. If the neuro-psychometric data also show an association between HFRT and better cognitive function, a further trial of HFRT might be discussed. Stratification of HFRT dose by age and biological risk factor would need to be guided by the present

Table 4 Secondary quality-of-survival outcomes by treat- ment group: Hormone replacement, use of therapy services, hearing aids, state benefits, and cosmetic outcome

n1, n2 (HFRT, STRT)

HFRT STRT

Outcome

Mean (SD) years to growth hormone replacement

27, 19 2.98 (0.7) 2.88 (0.6)

Growth hormone

72, 75 39 (54)

37 (49)

replacement therapy Thyroxine replacement therapy Occupational therapy Speech and language therapy Physiotherapy

71, 76 36 (51)

34 (45)

73, 77 46 (63) 73, 77 16 (22) 73, 77 26 (36)

56 (73) 15 (19) 23 (30)

Psychology

73, 77 29 (40) 72, 77 36 (50)

40 (52) 46 (60)

Special educational support

Educational provision not suited to child’s needs

60, 66 16 (27)

13 (20)

70, 74 16 (23) *

7 (10) *

Uses a hearing aid

In receipt of state benefits 71, 72 28 (40) Problems with appearance 71, 74 45 (63)

26 (36) 46 (62) 52 (73)

Hair thinning, patchy hair loss, or no hair

69, 71 58 (84)

Abbreviations as in Table 1 . Values are number (percentage) except where noted. Differences between treatment arms did not approach statistical significance ( P > .1) except where indicated. * c 2 4.81, P Z .028 for intergroup difference.