Key words
staging of cervical cancer - radiochemotherapy for cervical cancer - radical hysterectomy
- questionnaire on treatment situation - therapeutic concepts - stage-dependent therapy
Schlüsselwörter
Staging bei Zervixkarzinom - Radiochemotherapie bei Zervixkarzinom - radikale Hysterektomie
- Umfrage zur Behandlungssituation - Therapiekonzepte - stadienabhängige Therapie
Abbreviations
FIGO:
Fédération Internationale de Gynécologie et dʼObstétrique
GOG:
Gynecologic Oncology Group
NCCN:
National Comprehensive Cancer Network
OS:
Overall survival
PFS:
Progression-free survival
RCTX:
Radiochemotherapy
RH:
Radical hysterectomy
TMMR:
Total mesometrial resection
VALRH:
Vaginal-assisted laparoscopic radical hysterectomy
LARVH:
laparoscopic assisted radical vaginal hysterectomy
Introduction
The treatment of patients with invasive cervical cancer in Germany should be oriented
on the currently valid AGO guidelines [1]. These
guidelines, as well as international recommendations (e.g., NCCN) are based on
prospective and retrospective mono- and multicentric data, since only few
prospective randomised studies are available [2]. Even for
the radical hysterectomy (RH) which has been practiced for more than one hundred
years there are today still no randomised studies comparing the various surgical
techniques [99]. Also the classification of radicality
by various groups has not led to a standardisation of the surgical methods [3], [4]. In addition, because of
the declining incidence of invasive cervical cancer – with 4880 newly diagnosed
cases and 1600 deaths in Germany, fewer and fewer patients per hospital are being
treated which will lead to appreciable problems with regard to experience and
training [5].
On the other hand novel diagnostic [100], surgical and
radiooncological procedures are being introduced and need to be evaluated [101]. The staging system for cervical cancer according to
FIGO is today based solely on gynaecological examinations and cystoscopy/rectoscopy.
Although MRI and CT have clear limitations for the staging of women with cervical
cancer they are being employed more and more [6]. Whether
or not PET-CT can provide improved data for staging is currently not clear [7], [8], [9], [10]. Also the oncological relevance of
surgical staging is still being discussed controversially and is being checked in
the Uterus-11 study of AGO [11], [12], [13]. The sentinel concept appears to be
applicable with high sensitivity and detection rates for tumours < 2 cm; even so
there are reservations due to the low prevalence of positive lymph nodes in cases
of
tumours in stages IA1–IB1 [14], [15]. Beside the classical abdominal RH, in recent years other procedures
such as total laparoscopic RH [16], [17], [18], [19], [20], laparoscope-assisted radical
vaginal RH [21], [22], [23], [24] or vaginal-assisted
laparoscopic RH [25], robot-assisted RH [26], [27], [28] and TMMR [29] have been
developed and provided highly promising oncological results. Nerve-sparing
operations also help to markedly reduce the postoperative morbidity. Also from the
radiotherapy side, considerable improvements in the treatment of patients with
cervical cancer have been achieved. Beside the usual combination of chemotherapy and
radiation (radiochemotherapy = RCTX), which achieves a significant improvement in
survival for the patient [30], [31], [32], [33], [34], it is above all the use of
innovative techniques of radiation that clearly reduces the rate of side effects
caused by the therapy [35], [36]. The sole randomised study to compare radical hysterectomy with
radiotherapy alone by Landoni in 1997 must now be considered as outdated with regard
to the radiotherapy-associated side effects [37].
Also in the most recent publications on RH more than 50 % of the patients receive
a
postoperative adjuvant radiochemotherapy, which is often associated with an
increased morbidity [38], [39]. The criteria for an adjuvant therapy are applied in widely different
ways in spite of the GOG-109 data [40], [41], [42], [43]. Furthermore, different algorithms are used in the
follow-up [44], [45], [46].
The large number of open and controversial topics prompted our group to develop a
questionnaire to assess the current status in the treatment of women with cervical
cancer, even though we do not make any claims to completeness for this
compilation.
Material and Methods
Between February 2012 and June 2012 a questionnaire comprising 19 topic complexes
was
sent to the heads of all gynaecological hospitals nationwide in Germany
(Questionnaire – [Fig. 1]). The data were evaluated using
SPSS. Percentage values each refer to the number of usable replies to the particular
topic.
Fig. 1 Questionnaire.
Results
General
From 688 hospitals to which the questionnaire was sent 234 (34 %) answered, of
these 26 were university hospitals (11 %), 113 were medical centres (49 %) and
93 were regional and general hospitals (40 %). 28 % (n = 63) of these hospitals
are certified as centres for gynaecological oncology. All hospitals were
subdivided into 3 groups according to the number of patients with cervical
cancer treated per year: 1: 0–25 patients, 2: 26–50, 3: more than 50. The
university hospitals could be assigned as follows to the groups 1, 2 and 3,
respectively, 14 (61 %), 6 (26 %) and 3 (13 %). Three university hospitals did
not give an answer to the number of treated patients per year. For medical
centres and regional/general hospitals the assignments were as follows:
1 = 94 %, 2 = 6 %, 3 = 0 % and, respectively, 1 = 95.5 %, 2 = 4.5 % and
3 = 00 %. Altogether 91 % of all patients with cervical cancer were treated in a
hospital that handles less than 26 such patients per year.
Staging
In practically all hospitals, staging from stage IB1 was done according to FIGO
by bimanual examination (98 %). Cystoscopy and rectoscopy were done in 73 % and,
respectively, 70 % of the gynaecological departments. 44 % of the hospitals used
CT routinely, 52 % used MRI, whereas in only 2 % PET-CT was used for staging
([Fig. 2]). Surgical staging was preferred in
48 % of the responding hospitals. Here the following approaches for surgical
staging were used: open transperitoneal in 41 %, open retroperitoneal in 9 %,
laparoscopic transperitoneal in 47 %, laparoscopic extraperitoneal in 1 % and
robot-assisted transperitoneal in 2 %. The sentinel concept was not utilised in
43 %; 9 % of the hospitals performed sentinel lymphonodectomy in all patients
with tumours ≤ 2 cm, 22 % only within studies and 22 % only if requested by the
patient. In 9 gynaecological departments (4 %) the sentinel concept is applied
to all tumour stages ([Fig. 3]). Almost all hospitals
(96 %) reject the sentinel concept as an addition to conisation/simple
hysterectomy for patients with a cervical cancer stage pTIa1.
Fig. 2 Applied staging examinations from FIGO IB1 (multiple answers
possible).
Fig. 3 Use of the sentinel concept in patients with cervical
cancer.
Therapy
In the great majority of the gynaecological hospitals in Germany the classic open
Wertheim operation is used (71.5 %). Other procedures for RH with markedly lower
usage are represented by laparoscopic-assisted vaginal or vaginal-assisted
laparoscopic RH in 4 %, total laparoscopic RH in 10 %, TMMR in 13 %,
robot-assisted RH in 1 % and robot-assisted TMMR in 0.5 %. In 32 % of the
hospitals several procedures were regularly applied ([Fig. 4]). An intraoperative immediate section of all resected lymph
nodes is offered in 57 % of the hospitals, in 26 % only for suspiciously
enlarged lymph nodes. In 2 % of the hospitals – in the sense of a two-stage
procedure – a lymphonodectomy is performed first followed by RH only after
confirmation of tumour-free lymph nodes ([Fig. 5]).
Fig. 4 Methods for radical hysterectomy that were used predominantly
in the department.
Fig. 5 Intraoperative performance of an immediate section examination
of resected lymph nodes.
In the cases with the identification of positive pelvic lymph nodes in patients
with a local operable tumour the RH is discontinued in 16 % of the hospitals
and, after performance of a paraaortic lymphonodectomy, a primary
radiochemotherapy is initiated. 74 % of the physicians continue the RH
(including a paraaortic lymphonodectomy) and recommend adjuvant RCTX or adjuvant
chemotherapy (10 %) ([Fig. 6]). Upon identification
of positive paraaortic lymph nodes of a locally operable cervical carcinoma, the
operation is terminated in 50 % of the patients. In 43 % the RH is continued in
this clinical situation and an adjuvant RCTX is subsequently carried out, in 7 %
adjuvant chemotherapy ([Fig. 7]). Young patients with
a still unfulfilled desire to have children and a cervical cancer of less than
2 cm in size are referred in 80 % to a centre with expertise in radical
trachelectomy. 17 % of the hospitals perform uterus-sparing operations in house,
usually as a radical vaginal trachelectomy (14 %). Three percent of the
gynaecological departments reject a trachelectomy and always prefer a RH for
this constellation.
Fig. 6 Procedure in cases with positive pelvic lymph nodes on
immediate section.
Fig. 7 Procedure in cases with positive paraaortic lymph nodes on
immediate section.
In FIGO stage IB2 neoadjuvant chemotherapy is applied in 6 % of the clinics prior
to the planned RH and in 10 % a neoadjuvant RCTX, whereas in 75 % the RH is
performed as primary procedure. 9 % of the hospitals refer patients in this
stage to a primary RCTX. Also in FIGO stage IIB a primary RH followed by
adjuvant RCTX is preferred by the majority of German hospitals (46 %), primary
RCTX without and with laparoscopic staging in 21 % and 24 %, respectively. More
extensive therapeutic options are employed in 9 % of the hospitals ([Fig. 8]). For patients in stage IIIA/IIIB most
physicians consider a clinical staging to be sufficient prior to a primary RCTX
(69 %), merely 31 % vote for a surgical staging before commencing therapy. In
stage IVA of cervical cancer individual decisions are preferred in most
hospitals ([Fig. 9]).
Fig. 8 Preferred therapy for FIGO stage IIB.
Fig. 9 Primary therapy in stage IVA (multiple answers possible).
The indications for adjuvant RCTX after RH form a very heterogeneous pattern.
Whereas all hospitals (100 %) favour an adjuvant therapy in cases with positive
lymph nodes, invasion of the parametrium and an R1/R2 resection, the presence of
the risk factors stage 3, tumour size > 4 cm, age < 40 years,
adenocarcinoma as histological type and invasion of the lymphovascular space
(L1) alone or in combination is considered in widely differing ways ([Fig. 10] to [13]). In the
case of one of the above-mentioned risk factors 15-68% of the hospitals
recommend an adjuvant RCTX, in the case of 2 factors 42–88 %, for 3 factors
68–97 % and for 4 factors 90–97 %.
Fig. 10 Indications for adjuvant RCTX with one intermediate risk
factor.
Fig. 11 Indications for adjuvant RCTX with 2 intermediate risk
factors.
Fig. 12 Indications for adjuvant RCTX with 3 intermediate risk
factors.
Fig. 13 Indications for adjuvant RCTX with 4 intermediate risk
factors.
Follow-up
In the course of follow-up after primary radiochemotherapy in practically all
hospitals a gynaecological examination (99 %) and vaginal ultrasonography (99 %)
are performed. In 90 % of the hospitals in addition ultrasonography of the
kidneys is performed, a PAP smear in 75 %, a cervical abrasion in 28 %, a pelvic
MRI in 41 %, a tumour marker determination in 28 % and/or a PET-CT in 4 % ([Fig. 14]). On identification of a local persisting
tumour/intrauterine recurrence after primary radiochemotherapy only 28 % of the
hospitals consider a secondary hysterectomy as not being indicated, since it is
not associated with a benefit in terms of survival. In contrast 7 % of the
hospitals always perform a secondary hysterectomy, 57 % do so only when tumour
persistence is detected. Eight percent consider a primary exenteration to be
indicated in this clinical situation ([Fig. 15]).
Fig. 14 Examinations performed in the course of follow-up (multiple
answers possible).
Fig. 15 Indication for a secondary operation after primary
radiochemotherapy for cervical cancer.
Discussion
To what extent national guidelines are applied in the therapy for cervical cancer
can
only be determined on the basis of characteristic reimbursement data or with the
help of questionnaires [47], [48]. Whereas the former is only meaningful when a separate character can
be coded for every possible treatment modality, the latter is dependent on a high
participation. The response rate of 34 % achieved with the present questionnaire is
adequate and also allows representative conclusions on the composition of the
hospitals. The fact that within just a few years 28 % of all hospitals have attained
certification as oncological centres is indicative of an increased awareness of
quality in the treatment of patients with gynaecological cancers. In spite of the
gratifying decline in the incidence of these cancers, the number of patients with
cervical cancer treated per hospital must be considered very critically. The numbers
correspond to those gathered by Ackermann et al. in 2005 [98]. Only 19 hospitals of those that replied to our questionnaire treat
more than 25 patients per year when thereby all stages IA1 to IVA are collected
together. If this is extrapolated to all hospitals in Germany, at most 1000 patients
are treated in hospitals with extensive expertise and only very few in high-volume
hospitals. The great majority of women are treated in hospitals with low case
numbers which, in turn, will lead to increasing problems in further training and the
establishment of modern therapeutic concepts. The centralisation of therapy for
patients with invasive cervical cancer should be a target for the future.
In 2006 the FIGO committee again decided to leave the staging of cervical cancer as
a
purely clinical process [49]. All German hospitals confer
but only in 70 % are the cystoscopic and rectoscopic examinations required by FIGO
actually carried out. The routine use of CT and MRI in 44 % and 52 % reflect the
widely differing data with regard to the sensitivity and specificity of these two
imaging procedures found in the literature [50], [51], [52], [53], [54]. PET-CT, the costs of
which are not reimbursed in Germany, is of no significance as a staging method in
correspondence with its low sensitivity according to Ramirez et al. and LeBlanc et
al. [7], [8], [10], [55]. 48 % of the
hospitals consider surgical staging to be a valid alternative to imaging and to
clinical staging, although as yet no randomised clinical trial has been able to
confirm the oncological advantages seen in retro- and prospective studies [56], [57]. The successfully
recruiting prospective randomised Uterus-11 study of AGO on the value of surgical
staging in patients with cervical cancers of stages IIB–IVA should provide
pathbreaking answers to this matter. Why 41 % of the hospitals would still choose
an
open transperitoneal approach for surgical staging is not clear, since it is just
the formation of postoperative adhesions before an RCTX that is the main argument
against surgical staging.
The significance of the sentinel concept in cervical cancer is evaluated very
differently nationwide in Germany. The results of the Uterus-3 trial of AGO
published by Altgassen et al. still represent the largest investigation of this
topic worldwide. These results are suggestive that SLN can achieve a sufficiently
high detection and sensitivity in patients with a tumour size < 2 cm [14], [58]. Merely 9 % of all
hospitals in Germany employ the sentinel concept routinely for carcinomas < 2 cm.
44 % of the hospitals would apply the concept if requested by the patient or in
clinical trials. The international trend to define patient collectives who would
benefit from less radical surgery with equal oncological efficacy is only
implemented in a few hospitals [59]. In addition, the
advantages associated with the SLN technique such as the discovery of rare lymph
drainage pathways [60] or the detection of
micrometastasis [61] are not exploited. Lymph node
metastasis in stage IA1 is rare [15], [62]. This supports the opinion of the majority of
hospitals in Germany not to use SLN in this stage. It is, however, possible that by
this means one could identify with minimal morbidity just those few patients with
early lymph node metastasis, whose prognosis is otherwise rather poor [63], [64], [65].
The uterus-sparing operation in cases of early invasive cervical cancer is accepted
in 97 % of the gynaecological departments on account of its excellent oncological
and reproductive results [66], [67], [68]. In Germany this is almost
exclusively performed as radical vaginal trachelectomy in contrast to other
countries where abdominal radical trachelectomy also has a high relevance [47].
Various procedures are in use in Germany for radical hysterectomy, in some cases even
in one hospital. A positive development is that innovative procedures with excellent
oncological outcome, such as TMMR by the Leipzig group [29] or VALRH by the Berlin group [25] are
proportionally well represented. Thus we can hope that in the years to come the
monocentric data will be confirmed by multicentric results. Less easily
understandable are the answers of German hospitals in 74 % and 43 %, respectively,
to continue with RH in the presence of positive pelvic or paraaortic lymph nodes.
This could be based on the fact that 43 % of the hospitals do not perform
intraoperative immediate sections or on economic reasons, especially as practically
all hospitals recommend an adjuvant RCTX in the presence of positive lymph nodes.
The recommendation of GOG 109 study for an adjuvant therapy in cases of N1,
parametrial infiltration and an R1/R2 resection are almost unanimously implemented
in German hospitals [40], [69]. The comments of the gynaecological departments about an adjuvant therapy
after RH in the presence of only one or several intermediate risk factors clearly
reflect the, also internationally, much too high rate of trimodal therapy, the
oncological utility of which has not yet been clarified for many combinations of
these risk factors [41], [42].
Marnitz et al. have impressively demonstrated that the rate of adjuvant RCTX can be
minimised to 10 % with the help of laparoscopic staging and with knowledge of the
histological results of preoperative biopsies/conisation; in this way the
significantly elevated morbidity of RH and RCTX can be avoided [39]. For this complex of topics, in particular, a revised German
guideline with clear unambiguous statements is needed.
The optimal therapy for patients in FIGO stages IB2 and IIB has been a subject of
controversial discussion for many years, but is still not clear as is reflected in
the results of this and earlier questionnaires [98]. The
spectrum of nationally as well as internationally applied therapies encompasses
primary radical hysterectomy ± adjuvant radiochemotherapy, neoadjuvant
(radio)chemotherapy followed by radical hysterectomy, primary radiochemotherapy
(RCTX) or TMMR. In the not yet closed prospective randomised international EORTC
study 55994, neoadjuvant chemotherapy followed by radical hysterectomy is being
compared with primary RCTX in the tumour stages IB2–IIB. A randomised study to
compare the primary operation followed by adjuvant RCTX with primary RCTX for these
tumour stages has not yet been undertaken. In non-randomised studies that have
compared various therapeutic procedures, both significant and non-significant
differences have been presented [70], [71], [72], [73], [74], [75], [76], [77].
Whereas in the last century, an operation for patients with cervical cancers up to
stage IIB was in most cases the therapy of choice, in the present century primary
radiochemotherapy has become established worldwide for patients in FIGO stage IIB
on
the basis of the results of 5 prospective randomised studies. The significant
improvement in survival (referred to OS) in favour of combined radiochemotherapy as
compared to radiation alone has been proven in several metaanalyses. For patients
in
stage IIB the improvement in survival amounts to 7 % [33], or in a large retrospective analysis by Beck et al. on the basis of data
for 5476 patients even to 13 % [34]. The result of the
questionnaire with regard to therapy in stage IIB with a rate of 46 % for RH as
primary therapy is thus surprising.
For patients with a cervical cancer in stage IVA individual therapeutic decisions
should be made because there are no randomised comparisons between primary
exenteration and primary RCTX, this opinion is shared by 82 % of the German
hospitals. The question why only 13 % of the hospitals prefer a primary exenteration
in cases of pre-existing fistula formation must be evaluated critically with regard
to the available surgical expertise in Germany [78], [79], [80].
The follow-up of patients after primary therapy for cervical cancer, especially
primary RCTX, is regulated with regard to time intervals but not with regard to
extent. Also in Germany the clinical examination with vaginal ultrasonography is
used routinely even when its utility with regard to detection of local
recurrence/progression is limited. Thus, according to Duyn et al. only 32 % of the
recurrences are detected in the follow-up [81], or
according to Ansink et al. merely 26 % [82]. Imaging
procedures such as CT or MRI or PAP smears are, however, not superior to the
clinical examination [83]. The utility of MRI in the
follow-up is also disputed, as described by Balleyguier et al.[84], accordingly the moderate use in 41 % of the hospitals is justified.
PET-CT is associated with a high rate of false negative and false positive findings
in the diagnosis of recurrences as has been demonstrated by Chung et al. in 276
patients [85], whereas in other studies it has revealed a
good correlation between complete metabolic response and survival [86]. Further studies on the value of PET-CT in the
follow-up are needed. The determination of the tumour markers SCC and CEA is
undertaken in 28 % of the German hospitals [46], [87], even when the available data suggest that the
determination of these markers is not reliable [46], [87].
Routinely taken cytological samples – as is done in 75 % of the gynaecological
departments – should no longer be considered as the sole follow-up examination on
account of the low detection rate for recurrences of between 0 and 17 % [88], [89], [90], [91], [97]. According to Nijhuis et al. the combination of
examinations under anaesthesia with biopsy sampling/cervical abrasion can detect or
exclude a local persisting tumour after primary RCTX with a high probability [91]. This method should thus be employed more often than
its current implementation in merely 28 % of the hospitals.
The wide spectrum of answers with regard to secondary operation in patients after
primary RCTX for cervical cancer reflects the lack of evidence on this topic. While
Motton et al. did not observe any increased rate of complications, Classe et al. and
Colombo et al. recorded complications in 26–48 % of their secondary operations. To
what extent a simple hysterectomy is oncologically more meaningful than a radical
hysterectomy or an exenteration also remains to be clarified. None of the studies
could show any advantages with regard to overall survival, merely local control or,
respectively, PFS were improved [91], [92], [93], [94], [95], [96], [97]. A randomised study
of surgical strategies for persisting tumours is ethically not possible.
Of course, the results presented here have their limitations. The design and length
of the questionnaire did not allow the inclusion of all interesting questions and
all possible answers. Also the fact that only a good third of the hospitals
responded means that the results are still representative, although a higher
participation would have been highly desirable.
Conclusion
The results of this questionnaire on the therapy for cervical cancer in Germany in
2012 reveal on the one hand that innovative concepts (laparoscopic procedures for
RH, TMMR, surgical staging) have found acceptance in German hospitals while, on the
other hand, long established treatment concepts (RH in case of positive pelvic
and/or paraaortic lymph nodes, therapy in FIGO stage IIB, recommendation for
adjuvant therapy) are being retained. Randomised prospective studies should be put
into practice. The lack of evidence for many questions leaves room for a broad
spectrum of opinions and treatment pathways. The future new version of the S3
guidelines on the diagnostics and therapy for patients with cervical cancer should
thus include unambiguous statements for many of these aspects.
Acknowledgements
The authors are extremely grateful to all colleagues who returned the questionnaires
and provided helpful suggestions and support.
