39 / 1082
S26
ESTRO 36 2017
_______________________________________________________________________________________________
To provide a descriptive overview of the prospectively
collected physician assessed bowel morbidity and patient
reported outcome (PRO) within the multicenter EMBRACE
study, and to jointly evaluate the development of
individual symptoms.
Material and Methods
The analysis was based on 1419 patients enrolled from
2008-2015. Treatment included image guided adaptive
brachytherapy (IGABT) and EBRT delivered either by 3-
D conformal technique or IMRT/VMAT and chemotherapy.
Prescribed doses were 45-50Gy in 1.8-2.0Gy fractions. If
pathological lymph nodes were present, a boost to 55-
65Gy was given. Morbidity was assessed according to the
CTCAE v.3 and PRO according to EORTC QLQ C30/CX24 at
baseline, every 3 months (1
st
year), every 6 months (2
nd
and 3
rd
year) and yearly thereafter. Bowel endpoints
evaluated were diarrhea, flatulence, incontinence,
stenosis and fistula, all graded from 0 to 5 (flatulence G0-
G2). The related PRO was included with following
reportings; “a little”, “quite a bit” and “very much”.
Relevant cut-off values were applied to report CTCAE and
PRO: G≥2 and G≥1 versus “very much” and ≥”a little”.
Crude incidences, prevalences and actuarial estimates
were calculated.
Results
Baseline morbidity (BM) and follow up (FUP) information
was available for 1176 patients (PRO 942). Median follow
up was 27 months, 63% were treated with 3-D CRT and 37%
with IMRT/VMAT. Figure 1 illustrates the bowel symptoms
with prevalence rates at 5 years for diarrhea at 24% for
G≥1, and 4% for G≥2 (CTCAE). According to PRO, any
patient reported diarrhea was 35% and 3% for “very
much”. Both reached a plateau at a certain level during
FUP. Incontinence occurred in 9% as G≥1 and 2% as G≥2.
For PRO any patient reported difficulty in controlling
bowel was 29% and 3% for “very much”, both with
increasing prevalence during FUP. Crude incidences of
severe diarrhea and incontinence (G≥3, CTCAE) were 1.5%
and 0.4%, respectively. Sigmoid, small bowel and colon
stenosis G≥2 were present in 16 patients with 12 being
G3/G4 with only one morbidity-related death because of
necrotizing enteritis. Fistula G≥2 were present in 6
patients.
Crude incidences and actuarial estimates are
shown in table 1.
Conclusion
According to the data assessed within the EMBRACE study,
bowel morbidity is overall frequently reported, however,
severe morbidity is limited. The results indicate that
patients report higher burden of bowel symptoms,
however no direct correlation is possible between both
assessment methods. The challenge is to find a practical
way of interpreting the complementary information from
PRO regarding morbidity. The data illustrate that different
methodologies for quantification of morbidity provide
different results, with actuarial analysis indicating higher
magnitude than prevalence rates. In the future, better
methods for quantifying relevance and burden of
symptoms are warranted to further improve our morbidity
profile in cervix cancer patients.