International Journal of

Radiation Oncology
biology physics

www.redjournal.org

Clinical Investigation: Gynecologic Tumor

Manifestation Pattern of Early-Late Vaginal Morbidity After

Definitive Radiation (Chemo)Therapy and Image-Guided
Adaptive Brachytherapy for Locally Advanced Cervical
Cancer: An Analysis From the EMBRACE Study
Kathrin Kirchheiner, MSc,*,y Remi A. Nout, MD, PhD,z Kari Tanderup, PhD,x
Jacob C. Lindegaard, MD, DMSc,x Henrike Westerveld, MD, PhD,k
Christine Haie-Meder, MD,{ Primoz Petric, MD, PhD,#,**
Umesh Mahantshetty, DMRT, MD, DNB,yy Wolfgang Dörr, DVM, PhD,*,y
and Richard Pötter, MD, PhD*,y
*Department of Radiation Oncology, Comprehensive Cancer Center, Medical University of Vienna/General Hospital of
Vienna, Austria; yChristian Doppler Laboratory for Medical Radiation Research for Radiation Oncology, Medical University
of Vienna, Austria; zDepartment of Clinical Oncology, Leiden University Medical Center, The Netherlands; xDepartment of
Oncology, Aarhus University Hospital, Denmark; kDepartment of Radiotherapy, Academic Medical Centre, University of
Amsterdam, The Netherlands; {Department of Radiotherapy, Gustave-Roussy, Villejuif, France; #Department of
Radiotherapy, Institute of Oncology Ljubljana, Slovenia; **Department of Radiotherapy, National Center for Cancer Care
and Research, Doha, Qatar; and yyDepartment of Radiation Oncology, Tata Memorial Hospital, Mumbai, India

Received Dec 10, 2013, and in revised form Jan 14, 2014. Accepted for publication Jan 21, 2014.

Summary Background and Purpose: Brachytherapy in the treatment of locally advanced cervi-
Severe vaginal morbidity cal cancer has changed substantially because of the introduction of combined intraca-
within the first 2 years after vitary/interstitial applicators and an adaptive target concept, which is the focus of the
definitive radiation (chemo) prospective, multi-institutional EMBRACE study (www.embracestudy.dk) on image-
therapy including image- guided adaptive brachytherapy (IGABT). So far, little has been reported about the
guided adaptive brachyther- development of early to late vaginal morbidity in the frame of IGABT. Therefore,
apy with intracavitary/inter- the aim of the present EMBRACE analysis was to evaluate the manifestation pattern
stitial techniques for locally of vaginal morbidity during the first 2 years of follow-up.
advanced cervical cancer, is Methods and Materials: In total, 588 patients with a median follow-up time of
limited (grade 3, 3.6%) and 15 months and information on vaginal morbidity were included. Morbidity was pro-
is significantly less than spectively assessed at baseline, every 3 months during the first year, and every

Reprint requests to: Kathrin Kirchheiner, MSc, Department of Radiation Development and CIRROdthe Lundbeck Foundation Center for Investi-
Oncology, Comprehensive Cancer Center, Christian Doppler Laboratory for gational Research in Radiation Oncology. Sponsored by the Medical
Medical Radiation Research for Radiation Oncology, Medical University of University of Vienna. Research funds have been provided by Nucletron, an
Vienna, Austria, Waehringer Guertel 18-20, 1090 Vienna, Austria; Tel: Elekta company, and Varian Medical Systems.
þ(43) 1 40400 2639; E-mail: kathrin.kirchheiner@meduniwien.ac.at Conflict of interest: none.
Supported by the Austrian Federal Ministry of Economy, Family, and Supplementary material for this article can be found at
Youth and the Austrian Foundation for Research, Technology, and www.redjournal.org.

Int J Radiation Oncol Biol Phys, Vol. 89, No. 1, pp. 88e95, 2014
0360-3016/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ijrobp.2014.01.032
Volume 89  Number 1  2014 Vaginal morbidity after IGABT for cervical cancer 89

reported from earlier studies. 6 months in the second year according to the Common Terminology Criteria for
Thus, the new adaptive target Adverse Events, version 3, regarding vaginal stenosis, dryness, mucositis, bleeding,
concept seems to be a safe fistula, and other symptoms. Crude incidence rates, actuarial probabilities, and preva-
treatment with regard to the lence rates were analyzed.
vagina being an organ at risk. Results: At 2 years, the actuarial probability of severe vaginal morbidity (grade 3)
However, mild and moderate was 3.6%. However, mild and moderate vaginal symptoms were still pronounced
vaginal morbidity is still (grade 1, 89%; grade 2, 29%), of which the majority developed within 6 months.
pronounced (grade 1, 89%; Stenosis was most frequently observed, followed by vaginal dryness. Vaginal bleeding
grade 2, 29%) with and mucositis were mainly mild and infrequently reported.
currently applied image- Conclusion: Severe vaginal morbidity within the first 2 years after definitive radiation
guided adaptive brachyther- (chemo)therapy including IGABT with intracavitary/interstitial techniques for locally
apy, and it needs further advanced cervical cancer is limited and is significantly less than has been reported
attention. from earlier studies. Thus, the new adaptive target concept seems to be a safe treat-
ment with regard to the vagina being an organ at risk. However, mild to moderate
vaginal morbidity is still pronounced with currently applied IGABT, and it needs
further attention. Ó 2014 Elsevier Inc.

Introduction present analysis of the prospective multi-institutional

EMBRACE study (www.embracestudy.dk) was initiated.
The primary objective of this report is to evaluate early to
The state-of-the-art treatment for patients with locally
late vaginal morbidity during the first 2 years of follow-up.
advanced cervical cancer is definitive radiation (chemo)
This is the first prospective study in the frame of IGABT
therapy, with brachytherapy being an essential component
focusing on detailed evaluation of individual vaginal symp-
(1). In the past 2 decades, with the introduction of image-
guided adaptive brachytherapy (IGABT) based on toms with regard to their contribution to overall vaginal
morbidity and to the manifestation pattern of symptoms over
repeated volumetric imaging (computed tomography [CT],
time in a large, multi-institutional patient cohort.
magnetic resonance imaging [MRI]), local control has
substantially improved. This implies the detailed delinea-
tion of target volumes and organs at risk (OAR) and dose Methods and Materials
adaptation according to pre-established dose aims and
constraints, without prescribing the dose to the traditional The EMBRACE study is a prospective observational study with
point A but rather to a precisely specified target volume (2, 24 participating centers worldwide. It was approved in all
3). The overall incidence of (late) side effects shows also a participating centers by the respective National Ethics Com-
decreasing trend in comparison with the 2-dimensional mittees. Since July 2008, the study has included more than 1000
(2D) era (4-8). This new target concept, however, also patients and is about to finish accrual. For this report, patients
has a significant impact on the dose to some OAR (9), with data from at least 1 completed follow-up visit regarding
including the vagina, where the dose can vary substantially, morbidity and tumor control were included in the analysis.
with some patients receiving much less dose to the vagina
and some receiving considerably more (10). Large doses
are most notably present in patients with large tumors Patients
treated with combined intracavitary/transvaginal interstitial
brachytherapy. However, information about the clinical The inclusion criteria were a primary biopsy-proven squamous
consequences of these doses in the vagina as OAR and their carcinoma, adenocarcinoma, or adenosquamous carcinoma of
impact on the patient’s quality of life is very limited (11). the uterine cervix, International Federation of Gynecology and
Owing to the improvements in treatment outcome, a Obstetrics stage IB-IVA (and stage IVB with para-aortic
progressively increasing number of long-term survivors of metastatic nodes below L1-L2 only), treated with curative
cervical cancer is expected in the future. Severe treatment- intent with definitive radiation (chemo)therapy. The minimum
related side effects are rare, but mild to moderate morbidity staging requirements were gynecologic examination, MRI of
can have a significant impact on the patients’ well-being and the pelvis, abdominal CT or MRI, and chest radiography.
health-related quality of life (12). With regard to vaginal
morbidity, changes related to feminine self-perception and Treatment
sexual functioning are a particularly important source of
symptom-related distress in the years after treatment (13). All patients were treated by a combination of external beam
To characterize vaginal morbidity associated with current radiation therapy (EBRT) with or without chemotherapy and
treatment strategies for locally advanced cervical cancer, the IGABT, within a maximal overall treatment time of 50 days.
90 Kirchheiner et al. International Journal of Radiation Oncology  Biology  Physics

EBRT was delivered either by 3-D conformal radiation Actuarial incidences were calculated by Kaplan-Meier
therapy or by intensity modulated radiation therapy. The method, with end of treatment as the starting point. Events
clinical target volume (CTV) of the elective EBRT consisted were defined in patients with any vaginal morbidity (grade 1),
of the gross tumor volume (GTV), uterus, parametria, the in patients with vaginal morbidity grade2 and higher
proximal vagina at least 2 cm below the GTV, and at least all (grade 2), and in those with grade3 and higher (grade 3).
pelvic lymph node regions. The elective CTV received 45 to Ninety-five percent confidence intervals (CI) are given. In pa-
50 Gy in 1.8- to 2.0-Gy fractions, and the pathologic lymph tients with local and/or regional and/or systemic recurrence of
nodes received a boost to 60 to 65 Gy in 2.0- to 2.4-Gy disease, the vaginal morbidity was censored at the time of
fractions. IGABT was delivered either by pulsed dose rate or recurrence. Patients who were not in complete remission
high-dose-rate brachytherapy. MRI was required for the first directly after treatment but who had 3 consecutive follow-up
brachytherapy applicator implant; 3-dimensional (3D) im- investigations showing tumor control were kept in the analysis;
aging (MRI or CT) was required for each subsequent appli- patients with persistent disease were excluded from analysis.
cator implant, although more than 1 fraction could be In 162 patients, follow-up data of at least 24 months
delivered by the use of the same implant. In over 90% of were present for the analysis of manifestation pattern over
EMBRACE patients, MRI was used for each application. time. Individual symptoms were defined as persistent if
The target concept and dose prescription to the high-risk they were present with any grade 1 in at least 3 consec-
CTV (HR-CTV), intermediate-risk CTV (IR-CTV), and utive follow-up visits, or as fluctuating if they were present
OAR (bladder, rectum, sigmoid, and bowel) could be with any grade 1 but resolved to grade0 in a shorter time
chosen according to the institutional practice of the indi- (single and multiple occurrences possible). Patients were
vidual participating center but were reported in a uniform not evaluated in the analysis of manifestation pattern over
way according to the Gynaecological Groupe Européen de time if any grade 1 occurred in their last follow-up visit or
Curiethérapie - European Society for Therapeutic Radi- their last 2 follow-up visits.
ology and Oncology (2, 3). The SPSS statistical software system , version 17 (SPSS
Inc., Chicago, IL), was used for calculations.
Assessment of vaginal morbidity
Results
In the EMBRACE study, morbidity is assessed at baseline
and after the end of treatment every 3 months during the first At the date of closure of the database for this analysis in
year, every 6 months during the second and third years, and January 2013, 767 patients from 19 centers were included in
yearly thereafter up to 5 years. The Common Terminology the EMBRACE study. Follow-up data regarding morbidity
Criteria for Adverse Events, version 3 (CTCAE v3.0) (14), and tumor control were available in 588 patients, who were
were used to grade vaginal morbidity and included the thus included in this analysis. The median follow-up time was
following individual vaginal symptoms: dryness, stenosis, 15 months (range, 1-49 months), with 162 patients (28%)
mucositis, bleeding, fistula, and other (free text entry) (15). having at least 24 months of follow-up. The patient, disease,
An overview about the CTCAE v3.0 scoring is given in and treatment characteristics are reported in Table 1.
the supplementary material (table E1, available online at
www.redjournal.org). Each individual vaginal endpoint was Tumor-related individual vaginal symptoms before
scored prospectively at baseline and each follow-up visit by treatment
the treating physician.
The crude incidences of individual vaginal symptoms before
Endpoints and statistical evaluation treatment are reported in Table 2. In summary, vaginal dryness,
stenosis, and mucositis played a minor role before treatment
The crude incidences of the individual vaginal symptoms (grade0 incidence >90%). By contrast, tumor-related vaginal
were calculated with the maximum symptom grading per bleeding occurred in the majority of patients with 49% mild
patient over all follow-up visits. The crude incidence for the grade1, 15% moderate grade2, and 3% severe grade3. In 2
overall vaginal morbidity was calculated by use of the patients, grade4 vaginal bleeding with life-threatening con-
maximum symptom grading observed for any of the vaginal sequences was reported before treatment.
endpoints per patient over all follow-up visits.
The prevalence rates show the proportion of patients Overall vaginal morbidity (any individual vaginal
with individual vaginal symptoms in relation to all patients symptom)
at a certain follow-up time point. The prevalence rate for
overall vaginal morbidity was calculated by use of the Crude incidence rates of the overall vaginal morbidity (any
maximum grading observed in any of the vaginal morbidity individual vaginal symptom) showed that mild (grade1,
endpoints. The fixed time points for prevalence rates reflect 53%) to moderate (grade2, 19%) vaginal morbidity was
the scheduled follow-up of the EMBRACE web database frequently observed, the rate of any severe vaginal
and include certain variations. morbidity was low (grade3, 2%), and life-threatening
Volume 89  Number 1  2014 Vaginal morbidity after IGABT for cervical cancer 91

Table 1 Patient, disease and treatment characteristics consequences (grade4) occurred in only 1 patient, as re-
ported in Table 3.
Patient and disease characteristics (NZ588)
Age, y, median (range) 49 (22-91) The actuarial probability that a patient would have any
Missing, n (%) 23 (4%) vaginal morbidity grade 1 within 2 years after treatment
FIGO tumor stage was 89% (CI: 85.2-92.6), of which the majority occurred in
IB 105 (18%) the first 6 months after treatment. The probability for
IIA 41 (7%) grade 2 was estimated with 29% (CI: 23.6-33.4) and for
IIB 291 (50%)
IIIA 2 (<1%)
grade 3 with 3.6% (CI: 1.4-5.8). The prevalence rates for
IIIB 111 (19%) grade1 and grade2 increased during the first 2 years of
IVA 15 (3%) follow-up but seemed to reach a plateau at 2 years (Fig. 1).
IVB 8 (1%)
Tumor dimensions at MRI, mm
Mean width (SD) 48 (14) Treatment-related individual vaginal symptoms
Mean height (SD) 41 (15)
Mean thickness (SD) 40 (12)
Vaginal stenosis (shortening and/or tightening) was the
Missing, n (%) 22 (4%)
Histology most frequently observed individual vaginal symptom after
SQ 472 (80%) treatment, with a crude incidence of 59% for grade 1,
AC 74 (12%) 16% for grade 2, and 1% for grade 3 (Table 3). The
AdSQ 27 (5%) corresponding actuarial probability rates were 75% for
Vaginal involvement (MRI) grade 1 (CI: 70.1-79.5), 22% for grade 2 (CI: 17.0-
Not involved 304 (52%)
Upper third 236 (40%)
26.0), and 1.6% for grade 3 (CI: 0.2-3.0) at 2 years. The
Middle third 20 (3%) prevalence rates for vaginal stenosis grade1 and grade2
Lower third 10 (2%) increased continuously during the first 2 years of follow-up,
Treatment characteristics as shown in Figure 2A.
EBRT PTV-E total dose D95 in EQD2 Vaginal dryness was reported in nearly half of the pa-
Median dose, Gy (IQR) 45 (45-46)
tients, with a crude incidence of 47% for grade 1 but
Missing, n (%) 18 (3%)
EBRT technique
mainly in mild grade1 (42%); only 5% showed a moderate
3D conformal 449 (76%) grade2 vaginal dryness (Table 3). The actuarial probability
IMRT 121 (21%) for grade 1 dryness at 2 years was 62% (CI: 56.0-67.0),
Concomitant chemotherapy and for grade 2 it was 8% (CI: 4.9-10.7). The prevalence
No 30 (5%) rates for vaginal dryness showed a tendency of increasing
Yes 545 (93%)
grade1 over time, although not continuously (Fig. 2B).
BT HR-CTV D90 in EQD2
Median dose, Gy (IQR) 89 (84-94) Vaginal mucositis was seen in 29% for grade 1, in 4%
Missing, N (%) 36 (6%) for grade 2, and in <1% for grade 3 in crude incidence
Dose rate (Table 3), with actuarial probability of 37% (CI: 32.3-42.1),
HDR 307 (52%) 6% (CI: 3.4-8.2), and 0.9% (CI: 0.1-1.7), respectively. The
PDR 269 (46%) prevalence rates showed mainly mild grade1 mucositis,
Applicator
Ring 320 (55%)
which seemed to decrease slightly over time (Fig. 3A).
Ovoids 165 (28%) Vaginal bleeding after treatment showed low prevalence
Molds 72 (12%) rates on a mild level (mainly grade1) and remained quite
Other 20 (3%) stable over time. The crude incidence rate of vaginal
Application bleeding was 31% for grade 1, 1% for grade 2, and
Intracavitary 399 (68%)
<1% for grade 3, with actuarial probability for grade 1
Intracavitary/interstitial 177 (30%)
Abbreviations: AC Z adenocarcinoma; AdSQ Z adenosquamous
carcinoma; BT Z brachytherapy; EBRT Z external beam radiation Table 2 Crude incidences of individual vaginal symptoms
therapy; EQD2 Z equivalent dose in 2-Gy fractions;
before treatment (NZ588)
FIGO Z International Federation of Gynecology and Obstetrics;
HDR Z high-dose-rate; HR-CTV D90 Z high-risk clinical target Vaginal Vaginal Vaginal Vaginal
volume receiving 90% of the prescribed dose; IMRT Z intensity Grade stenosis dryness mucositis bleeding
modulated radiation therapy; IQR Z interquartile range;
G0 560 (95%) 547 (93%) 533 (91%) 174 (30%)
MRI Z magnetic resonance imaging; N Z number of total;
PDR Z pulsed-dose-rate; PTV-E total dose D95 Z planning target G1 6 (1%) 21 (4%) 34 (6%) 291 (49%)
volume of elective field receiving 95% of the prescribed dose; G2 4 (1%) 2 3 (1%) 86 (15%)
SD Z standard deviation; SQ Z squamous carcinoma; WHO Z World G3 0 N.A. 0 17 (3%)
Health Organization. G4 N.A. N.A. 0 2
Proportions are given with number of total and percentages; missing G5 N.A. N.A. 0 0
data are in the range between 2% and 4%. Continuous variables are Missing 18 (3%) 18 (3%) 18 (3%) 18 (3%)
given with mean/median and standard deviation/range or interquartile
Abbreviation: N.A. Z not applicable.
range; missing data are declared.
92 Kirchheiner et al. International Journal of Radiation Oncology  Biology  Physics

Table 3 Crude incidences of treatment-related individual vaginal symptoms and overall vaginal morbidity in 588 patients with a
median follow-up time of 15 months
Vaginal Vaginal Vaginal Vaginal Vaginal Other vaginal Overall vaginal
Grade stenosis dryness mucositis bleeding fistula symptoms morbidity
G0 241 (41%) 312 (53%) 415 (71%) 407 (69%) 582 (99%) 523 (89%) 155 (26%)
G1 256 (43%) 244 (42%) 146 (25%) 175 (30%) 2 47 (8%) 309 (53%)
G2 86 (15%) 32 (5%) 23 (4%) 5 (1%) 0 14 (2%) 111 (19%)
G3 5 (1%) N.A. 3 1 4 (1%) 4 (1%) 12 (2%)
G4 N.A. N.A. 1 0 0 0 1
G5 N.A. N.A. 0 0 0 0 0
Abbreviation: N.A. Z not applicable.

at 2 years of 40% (CI: 35.3-45.5), for grade 2 of 2% (CI: prospectively followed up. The aim of this analysis was to
0-3.6), and for grade 3 of 0.3% (CI: 0-0.9) (Table 3, describe vaginal morbidity within the first 24 months after
Fig. 3B). the end of treatment.
Other vaginal symptoms occurred in 11% for grade 1, Two years after treatment, there was a low incidence of
in 3% for grade 2, and in 1% for grade 3 in crude any serious vaginal side effects (grade3 and grade4, 2%),
incidence (Table 3); most frequently reported in the free but a high actuarial probability that a patient would have
text field were vaginal discharge, adhesions, dyspareunia, any grade1 or higher vaginal morbidity (89%), of which the
and telangiectasia. majority occurred already within the first 6 months after
treatment. Vaginal stenosis was most frequent, with the
Severe vaginal morbidity highest actuarial probability rates both for grade 1 (75%)
and for grade 2 (22%), followed by vaginal dryness. The
prevalence rates of both vaginal stenosis and dryness
In 12 patients, 17 severe grade3 symptoms occurred, and in
increased during follow-up. By contrast, vaginal bleeding
1 patient a grade4 life-threatening symptom was reported
(grade4 mucositis, tissue necrosis with significant sponta-
neous bleeding and life-threatening consequences). Five
complete obliterations of the vagina (grade3 stenosis) were
reported. Three confluent ulcerations, bleeding with
trauma, and inability to tolerate vaginal examination, sexual
intercourse, or tampon placement (grade3 mucositis) were
observed. Vaginal bleeding, indicating interventional
treatment (grade3 bleeding) occurred once. Four symp-
tomatic fistulas were reported, which interfered with ac-
tivities of daily life with indication for invasive intervention
(grade3 fistula); 3 of them were vesicovaginal fistulas, and
1 was rectovesicovaginal. Four severe (grade3) other
vaginal symptoms were observed in the free text field: 2
ulcerations and 2 fibroses.

Analysis of manifestation pattern of individual

symptoms over time

The evaluation during the first 24 months of follow-up

revealed that vaginal stenosis and dryness were persistent
during at least 3 consecutive follow-up points in 62% and
55% of the patients, and mucositis and bleeding fluctuated
in 82% and 75% of the patients (Table 4).

Discussion Fig. 1. Overall vaginal morbidity (all individual vaginal

symptoms combined). Prevalence rates for all gradings at
The ongoing EMBRACE study represents a large cohort of each follow-up visit are shown in bar charts. Actuarial
patients with locally advanced cervical cancer treated with incidences for grade 1, grade 2, and grade 3 are
definitive radiation (chemo)therapy and IGABT who were shown in Kaplan-Meier curves.
Volume 89  Number 1  2014 Vaginal morbidity after IGABT for cervical cancer 93

Fig. 2. (A) Vaginal stenosis. Prevalence rates for all gradings at each follow-up visit are shown in bar charts. Actuarial
incidences for grade 1, grade 2, and grade 3 are shown in Kaplan-Meier curves. (B) Vaginal dryness. Prevalence rates
for all gradings at each follow-up visit are shown in bar charts. Actuarial incidences for grade 1 and grade 2 are shown in
Kaplan-Meier curves.

decreased substantially after treatment and, together with planning from the NOCECA study (Nordic protocol for CT
mucositis, remained stable at low levels during follow-up. based EBRT in locally advanced cervical cancer) (1994-
The overall low rate of severe but frequent rate of mild 2000). As in the STIC trial, a significant reduction of
to moderate vaginal morbidity is in line with other studies vaginal morbidity was reported with 3D treatment plan-
of the use of definitive radiation (chemo)therapy with ning, both for grade 2 (2D, 87%, vs 3D, 33%) and for
IGABT. In the prospective French STIC study (Programme grade 3 (2D, 9%, vs 3D, 4%) (7).
de Soutien aux Techniques Innovantes Coûteuses) The Vienna IGABT experience provides data with a
comprising 3 subgroupsd2 subgroups with neoadjuvant median follow-up time of 48 months (5). With this pro-
radiation therapy and 1 subgroup with definite radiation longed follow-up time, only 15% of patients remained
(chemo)therapyda comparison was reported between pa- without any vaginal morbidity. The crude rate of moder-
tients treated with 2D and 3D brachytherapy treatment ate (grade2, 28%) morbidity was slightly higher, and
planning (based on CT). With a median follow-up time of grade3 or grade4 (2%) are in accordance with our results.
24 months, significantly fewer overall gynecologic side With additional follow-up in the present study, it is ex-
effects according to CTCAE v.3 grade 3/4 were observed in pected that the findings will be further substantiated and
the subgroup of 117 patients treated with definitive that analysis of this large prospective multi-institutional
radiation-(chemo)therapy and 3D-IGABT (actuarial rate of EMBRACE study will provide more insight into the
2D, 15.4%, vs 3D, 1.4%) (6). question whether the overall vaginal morbidity and/or the
In addition, Lindegaard et al (7) reported recently on 140 individual symptoms will gradually increase or stabilize
patients treated in 1 institution (Aarhus) with definitive over time.
radiation (chemo)therapy plus IGABT. Quite in line with Historically, in the time before the integration of MRI into
the presented results, the authors found a 3-year actuarial treatment planning, the same institution reported a 3-year
rate of 33% for overall vaginal morbidity grade 2 and 4% actuarial rate of overall vaginal morbidity grade 3 of 30.6%
for grade 3 according to the CTCAE v.3. (mainly shortening and obliteration) (16). Although
In the same publication (7), the outcome in patients morbidity was assessed with the Late Effects Normal Tis-
treated with IGABT was historically compared with that in suedSubjective, Objective, Management, Analytic scales,
99 patients treated with radiographic brachytherapy which are not entirely comparable with CTCAE v.3, the
94 Kirchheiner et al. International Journal of Radiation Oncology  Biology  Physics

Fig. 3. (A) Vaginal mucositis. Prevalence rates for all gradings at each follow-up visit are shown in bar charts. Actuarial
incidences for grade 1, grade 2, and grade 3 are shown in Kaplan-Meier curves. (B) Vaginal bleeding. Prevalence rates
for all gradings at each follow-up visit are shown in bar charts. Actuarial incidences for grade 1, grade 2, and grade 3
are shown in Kaplan-Meier curves.

results underline the major reduction in vaginal morbidity depending on patients’ sexual activity (ie , grade2 vaginal
resulting from image-guided adaptive treatment planning. dryness is defined as “interfering with function, dyspar-
There are several limitations of this report. The eunia, intervention indicated”), reflecting a subjective
EMBRACE study is not a randomized study but an evaluation of dryness in sexually active patients. There
observational trial. Therefore, outcome can be compared exists no severe (grade3) or life-threatening (grade4)
with the literature and with historical data. vaginal dryness. A moderate grade2 stenosis follows the
Although the grading of severe or life-threatening definition of “vaginal narrowing and/or shortening inter-
symptoms (grade 3/4) is clearly described in CTCAE fering with function,” whereas it remains unclear whether
v3.0, mainly in conjunction with the required medical the interference with function is related to intercourse, to
intervention, the grading of mild to moderate symptoms dilation, or to the gynecologic examination as well.
(grade 1/2) is subject to subjective interpretation. It has to Therefore, both endpoints do not clearly reflect objectively
be critically questioned whether certain variations in the assessed vaginal changes but take additionally subjective
range of milder grading are due to healing effects or rather evaluations into account. It has been shown previously that
reflect interobserver variations between assessing physi- physician-assessed and patient-reported symptoms are not
cians. Furthermore, vaginal dryness and stenosis are graded well in accordance (17). In this EMBRACE report,

Table 4 Analysis of manifestation pattern of individual symptoms over time in a subgroup of 162 patients with at least 24 months of
follow-up
Symptoms with manifestation Vaginal stenosis, Vaginal dryness, Vaginal mucositis, Vaginal bleeding,
pattern over time nZ114 nZ84 nZ55 nZ68
Fluctuating symptoms 43 (38%) 38 (45%) 45 (82%) 51 (75%)
Persisting symptoms 71 (62%) 46 (55%) 10 (18%) 17 (25%)
Patients were excluded from this analysis if no grade 1 symptoms were recorded in follow-up visits or if any grade 1 symptom occurred in their last
follow-up visit or their last 2 follow-up visits.
Persistent symptoms are defined, if present with any grades 1 in at least 3 consecutive follow-up visits. Fluctuating symptoms are defined, if present
with any grades 1, but resolved to grade0 in a shorter period of time (single and multiple occurrences possible).
Volume 89  Number 1  2014 Vaginal morbidity after IGABT for cervical cancer 95

patient-reported symptoms have not been included, which and terms in 3D image based 3D treatment planning in cervix cancer
limits the informative value. brachytherapy with emphasis on MRI assessment of GTV and CTV.
Radiother Oncol 2005;74:235-245.
With regard to the reporting of morbidity endpoints, 3. Pötter R, Haie-Meder C, Van Limbergen E, et al. GEC ESTRO
crude incidences tend to underestimate morbidity outcome Working Group. Recommendations from Gynaecological (GYN) GEC
because they do not take into account the increasing ESTRO working group (II): Concepts and terms in 3D image-based
probability of patients remaining at risk. By contrast, treatment planning in cervix cancer brachytherapy-3D dose volume
actuarial incidences bear the risk of overestimating parameters and aspects of 3D image-based anatomy, radiation physics,
radiobiology. Radiother Oncol 2006;78:67-77.
morbidity because symptoms are assumed to be permanent 4. Pötter R, Dimopoulos J, Georg P, et al. Clinical impact of MRI
after the first occurrence (18). Especially in the analysis of assisted dose volume adaptation and dose escalation in brachytherapy
early to late morbidity occurring within the first 2 years of locally advanced cervix cancer. Radiother Oncol 2007;83:148-155.
after treatment, healing effects are not taken into account in 5. Pötter R, Georg P, Dimopoulos JC, et al. Clinical outcome of protocol
the Kaplan-Meier estimate. According to the present ob- based image (MRI) guided adaptive brachytherapy combined with 3D
conformal radiotherapy with or without chemotherapy in patients with
servations of the manifestation pattern of individual locally advanced cervical cancer. Radiother Oncol 2011;100:116-123.
symptoms over time, several vaginal symptoms tend to 6. Charra-Brunaud C, Harter V, Delannes M, et al. Impact of 3D image-
fluctuate over time; especially regarding mild mucositis and based PDR brachytherapy on outcome of patients treated for cervix
bleeding, the majority of patients showed a transient carcinoma in France: Results of the French STIC prospective study.
occurrence, meaning that symptoms had healed and dis- Radiother Oncol 2012;103:305-313.
7. Lindegaard JC, Fokdal LU, Nielsen SK, et al. MRI-guided adaptive
appeared by 1 of the next follow-up examinations. Vaginal radiotherapy in locally advanced cervical cancer from a Nordic
stenosis, by contrast, tends to display persistence or pro- perspective. Acta Oncol 2013;52:1510-1519.
gression over time and thus reflects chronic and irreversible 8. Vale C, Tierney JF, Stewart LA, et al. Chemoradiotherapy for Cervical
tissue changes, likely mainly as a result of increased Cancer Meta-Analysis Collaboration. Reducing uncertainties about
collagen production within the fibroconnective tissue. the effects of chemoradiotherapy for cervical cancer: A systematic
review and meta-analysis of individual patient data from 18 random-
Future EMBRACE analyses will focus on patient- ized trials. J Clin Oncol 2008;26:5802-5812.
reported vaginal symptoms and analyze their impact on 9. Tanderup K, Nielsen SK, Nyvang GB, et al. From point A to the
sexuality. Furthermore, detailed patient-related, disease- sculpted pear: MR image guidance significantly improves tumour dose
related, and treatment-related (especially dosimetric) pa- and sparing of organs at risk in brachytherapy of cervical cancer.
rameters will be evaluated with regard to vaginal morbidity. Radiother Oncol 2010;94:173-180.
10. Westerveld H, Pötter R, Berger D, et al. Vaginal dose point reporting
in cervical cancer patients treated with combined 2D/3D external
Conclusion beam radiotherapy and 2D/3D brachytherapy. Radiother Oncol 2013;
107:99-105.
11. Fidarova EF, Berger D, Schüssler S, et al. Dose volume parameter
Severe grade 3/4 vaginal morbidity after definitive radiation D2cc does not correlate with vaginal side effects in individual patients
(chemo)therapy including IGABT for locally advanced with cervical cancer treated within a defined treatment protocol with
cervical cancer within 2 years is limited and significantly very high brachytherapy doses. Radiother Oncol 2010;97:76-79.
12. Bergmark K, Avall-Lundqvist E, Dickman PW, et al. Vaginal changes
less than in previous studies with radiographic treatment
and sexuality in women with a history of cervical cancer. N Engl J
planning. Thus, the new adaptive target concept including Med 1999;340:1383-1389.
the use of intracavitary/interstitial techniques seems to be a 13. Juraskova I, Butow P, Robertson R, et al. Post-treatment sexual
safe treatment with regard to major morbidity in the vagina adjustment following cervical and endometrial cancer: A qualitative
being an OAR. However, mild to moderate morbidity oc- insight. Psychooncology 2003;12:267-279.
14. Cancer Therapy Evaluation Program, Common Terminology Criteria
curs in the majority of patients within the first 2 years after
for Adverse Events, Version 3.0, DCTD, NCI, NIH, DHHS March 31,
the end of treatment and is still a substantial problem, 2003 (http://ctep.cancer.gov). Publish date: August 9, 2006.
which deserves further attention. Mild vaginal symptoms 15. Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: Development of a
are not always persistent but can heal and fluctuate over comprehensive grading system for the adverse effects of cancer
time. Vaginal stenosis (shortening and/or tightening) and treatment. Semin Radiat Oncol 2003;13:176-181.
16. Pötter R, Knocke TH, Fellner C, et al. Definitive radiotherapy based
dryness are most frequent and tend to be more persistent
on HDR brachytherapy with iridium 192 in uterine cervix carcinoma:
over time. Report on the Vienna University Hospital findings (1993-1997)
compared to the preceding period in the context of ICRU 38 recom-
mendations. Cancer Radiother 2000;4:159-172.
References 17. Kirchheiner K, Nout R, Lindegaard J, et al. Do clinicians and patients
agree regarding symptoms? A comparison after definitive radio-
1. Barbera L, Thomas G. Management of early and locally advanced chemotherapy in 223 uterine cervical cancer patients. Strahlenther
cervical cancer. Semin Oncol 2009;36:155-169. Onkol 2012;188:933-939.
2. Haie-Meder C, Pötter R, Van Limbergen E, et al. Gynaecological 18. Bentzen SM, Vaeth M, Pedersen DE, et al. Why actuarial estimates
(GYN) GEC-ESTRO Working Group. Recommendations from should be used in reporting late normal-tissue effects of cancer
Gynaecological (GYN) GEC-ESTRO Working Group (I): Concepts treatment. now!. Int J Radiat Oncol Biol Phys 1995;32:1531-1534.