Recommendations for detection in primary care are flawedBMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d3022 (Published 17 May 2011) Cite this as: BMJ 2011;342:d3022
All rapid responses
The problem with this guidance is that it does not have the
perspective of a general practitioner faced with the diagnostic difficulty
of pelvic/lower abdominal symptoms. Ovarian cancer is only one
possibility. Only excluding ovarian cancer is not the aim of the
diagnostic process.The guidelines are primarily interested in diagnosing
ovarian cancer; but this is not how patients present.
Ultrasound as first line is commonly preferred because it is a better
and broader diagnostic tool. The guidelines give no evidence suggesting
that this is bettered by a Ca-125 for women presenting with symptoms which
could be ovarian cancer - or many other things. By becoming so centred on
ovarian cancer, we risk losing our perspective as truly holistic doctors.
Competing interests: No competing interests
1 Department of Obstetrics and Gynaecology, Imperial College London, Queen Charlotte's and Chelsea Hospital, London, UK; 2 Department of Obstetrics and Gynaecology, University Hospitals K.U.Leuven, Leuven, Belgium; 3 Department of Electrical Engineering, Katholieke Universiteit Leuven, Leuven, Belgium; 4 Leuvens Kankerinstituut, University Hospitals K.U.Leuven, Leuven, Belgium; 5 Oncogynaecological Centre - Department of Obstetrics and Gynaecology, General Teaching Hospital of Charles University, Prague, Czech Republic
Correspondence to: Dirk Timmerman, firstname.lastname@example.org
Detecting ovarian cancer in primary care is challenging because of the nonspecific nature of the symptoms. Furthermore, characterising ovarian pathology once women have reached secondary care is fundamental to optimising management. Inappropriate surgery can worsen prognosis, and it is crucial for optimal patient outcome and cost-effectiveness to treat patients with suspected ovarian cancer in specialised units with a multidisciplinary approach [1-3]. Given these issues, the recent NICE guidelines are timely and had the potential to make an important difference to women both with a benign ovarian cyst as well as those with an ovarian cancer [4,5]. However, we believe that this is an opportunity missed. In our comments we focus on the main recommendations in relation to diagnosis (Table 1).
First tests for diagnosis in primary care
These recommendations will result in many women undergoing unnecessary investigation. A large number of premenopausal women with either symptoms of possible ovarian cancer or an ovarian cyst will have a serum CA125 above the suggested threshold of 35 IU/ml due to the lack of specificity of this marker in younger women. Unfortunately, a CA125 of less than 35 IU/ml is not necessarily reassuring, the sensitivityof this test for early stage disease is such that approximately 40% of early stage cancers will be missed if this protocol is followed. For example, data from the International Ovarian Tumour Analysis (IOTA) group show that 63/159 patients with stage I invasive cancer had a CA125 <35 (95% CI 32%-47%).
Furthermore, a large number of ultrasound scans will also be scheduled as a second stage test if these recommendations are followed, and it is highly likely that this will lead to unnecessary intervention in some women, especially as the guidelines do not make clear the ultrasound criteria they recommend to characterise an adnexal mass once one has been found. Therefore, the evidence suggests that ultrasonography would be a better first line investigation in women with symptoms suggestive of ovarian cancer. External validation of ultrasound-based simple rules and risk prediction models have confirmed the excellent performance of ultrasound to distinguish between benign and malignant ovarian masses [6,7]. If ultrasonography is appropriately used, this could lead to a reduction of costs for the National Health Service (NHS) and reduced anxiety in patients confronted with an abnormal test that they may believe indicates the likely presence of ovarian cancer.
Tumor markers to help establish the diagnosis in secondary care
In a large study from the IOTA group, we have demonstrated that CA125 offers no benefit when an ultrasound scan can be performed by an appropriately trained operator [8,9]. Furthermore, we also know that this marker does not improve the performance of the mathematical models and simple rules that can be used to characterise ovarian pathology [6,10]. Against this evidence base, the recommendation to measure CA125 seems unjustified. The IOTA group has also published on the morphological features of germ cell tumours and it should be possible to be more selective regarding when or if to measure tumour markers to help establish such a diagnosis. In secondary care, we believe ultrasound examination by an experienced operator should be the first line investigation, because this is the best available test to distinguish between most benign and malignant adnexal masses.
Malignancy indices to help establish the diagnosis in secondary care
The RMI  is a reasonable model, however it performs significantly less well than an ultrasound scan by an experienced operator and so should not be the test of choice in the secondary care setting . The selection of an appropriate cut-off value for RMI is challenging and currently the RMI is not calibrated to give clinicians an absolute risk of malignancy. Determining the optimal RMI threshold will not influence overall test performance but merely reflects the balance between false positives and false negatives. Whilst the RMI appears to have acceptable test performance for many masses, when applied to masses in young women and pathology that is difficult to characterise with ultrasound, the performance is poor [6,9,10]. In addition, the RMI has a relatively low sensitivity for early stage disease. On the other hand, the ultrasound-based simple rules and other prediction models have been shown to have significantly better overall test performance [6,10].
There is value in trying to use diagnostic tests to improve patient management in women with suspected ovarian cancer. Optimal clinical management is different if a mass is known to be borderline, metastatic, or invasive. Prediction models now exist that can distinguish between benign, borderline, and invasive tumours with high accuracy, and this is the challenge in modern diagnostics in relation to improving patient care . Future research should focus on the relative merits of ultrasonography, CT scan, and MRI for classifying both the subtype and for staging of ovarian cancer.
Procedures for imaging in the diagnostic pathway
We would agree that an ultrasound scan of the pelvis should be the first line test for any woman with possible ovarian cancer. This should be performed transvaginally. In the event that this scan is suggestive of malignancy, then an abdominal scan should be performed in order to assess the extent of metastatic disease. It is disappointing that NICE has not commented on the ultrasound markers that are important when discriminating between benign and malignant pathology [6,14,15]. There are straightforward simple rules that can be applied on the majority of ovarian masses which characterise ovarian pathology with a high level of test performance . The best test for any mass where simple rules do not apply is an ultrasound scan by an experienced operator. In addition, mathematical models have been developed that distinguish between benign and malignant masses. These have undergone temporal and external validation that confirm their robust performance [7,12].
The recommendations from NICE appear to be based on a single meta-analysis  and a small study of RMI . The latter study's results are inconsistent with previous validations of the RMI as adjusted by Tingulstad and colleagues, reporting nearly perfect sensitivity and specificity values whereas the meta-analysis reports clearly lower figures. It is unfortunate that they have not taken the opportunity to incorporate some of the largest studies in the literature into their guidance.
A further concern relates to the confusion in the guidelines in relation to second stage tests. They refer to tests used in the context of an ovarian cancer screening trial and extrapolate this to masses found in women presenting to primary or secondary care with symptoms. Test performance may be completely different in these two populations and such an approach is not appropriate.
The management of ovarian pathology is a common problem in gynaecology. Only a very small number of ovarian cysts are malignant and inappropriate referral to secondary care, unnecessary surgery or overly invasive surgical intervention are all significant risks to patients with a cyst that is inappropriately characterised . Most either require no intervention at all, or can be managed laparoscopically with relatively little inconvenience to the patient. The NICE guidance as presented will lead to many women either being unnecessarily referred to a hospital or undergoing procedures they do not require.
1 Vergote I, De Brabanter J, Fyles A, Bertelsen K, Einhorn N, Sevelda P, et al. Prognostic importance of degree of differentiation and cyst rupture in stage I invasive epithelial ovarian carcinoma. Lancet 2001;357:176-82.
2 Bristow RE, Berek JS. Surgery for ovarian cancer: how to improve survival. Lancet 2006;367:1558-60.
3 Greving JP, Vernooij F, Heintz APM, van der Graaf Y, Buskens E. Is centralization of ovarian cancer care warranted? A cost-effectiveness analysis. Gynecol Oncol 2009;113:68-74.
4 National Institute for Health and Clinical Excellence. Recognition and initial management of ovarian cancer . (Clinical guideline 122.) 2011. www.nice.org.uk/CG122.
5 Redman C, Duffy S, Bromham N, Francis K. Recognition and initial management of ovarian cancer: summary of NICE guidance. BMJ 2011;342:d2073.
6 Timmerman D, Ameye L, Fischerova D, Epstein E, Melis GB, Guerriero S, et al. Simple ultrasound rules to distinguish between benign and malignant adnexal masses before surgery: prospective validation by IOTA group. BMJ 2010;341:c6839.
7 Timmerman D, Van Calster B, Testa AC, Guerriero S, Fischerova D, Lissoni AA, et al. Ovarian cancer prediction by logistic regression models: a prospective evaluation of diagnostic accuracy. Ultrasound Obstet Gynecol 2010;36:226-234.
8 Van Calster B, Timmerman D, Bourne T, Testa AC, Van Holsbeke C, Domali E, et al. Discrimination between benign and malignant adnexal masses by specialist ultrasound examination versus serum CA-125. J Natl Cancer Inst 2007;99:1706-1714.
9 Valentin L, Jurkovic D, Van Calster B, Testa A, Van Holsbeke C, Bourne T, et al. Adding a single CA 125 measurement to ultrasound imaging performed by an experienced examiner does not improve preoperative discrimination between benign and malignant adnexal masses. Ultrasound Obstet Gynecol 2009;34:345-354.
10 Timmerman D, Van Calster B, Jurkovic D, Valentin L, Testa AC, Bernard JP, et al. Inclusion of CA-125 does not improve mathematical models developed to distinguish between benign and malignant adnexal tumors. J Clin Oncol 2007;25:4194-4200.
11 Jacobs I, Oram D, Fairbanks J, Turner J, Frost C, Grudzinskas JG. A risk of malignancy index incorporating CA 125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer. Br J Obstet Gynaecol 1990;97:922-929.
12 Van Holsbeke C, Van Calster B, Testa AC, Domali E, Lu C, Van Huffel S, et al. Predicting malignancy of ovarian tumors: prospective evaluation of models from the IOTA study. Clin Cancer Res 2009;15:684-691.
13 Van Calster B, Valentin L, Van Holsbeke C, Testa AC, Bourne T, Van Huffel S, et al. Polytomous diagnosis of ovarian tumors as benign, borderline, primary invasive or metastatic: development and validation of standard and kernel-based risk prediction models. BMC Med Res Methodol 2010;10:96.
14 Hennessy BT, Coleman RL, Markman M. Ovarian cancer. Lancet 2009;374:1371-1382.
15 Timmerman D, Testa AC, Bourne T, Ferrazzi E, Ameye L, Konstantinovic ML, et al. A logistic regression model to distinguish between the benign and malignant adnexal mass before surgery: a multicenter study by the International Ovarian Tumor Analysis (IOTA) group. J Clin Oncol 2005;23:8794-8801.
16 Geomini P, Kruitwagen R, Bremer GL, Cnossen J, Mol BWJ. The accuracy of risk scores in predicting ovarian malignancy. A systematic review. Obstet Gynecol 2009;113:384-394.
17 Raza A, Mould T, Wilson M, Burnell M, Bernhardt L. Increasing the effectiveness of referral of ovarian masses from cancer unit to cancer center by using a higher referral value of the risk of malignancy index. Int J Gynecol Cancer 2010;20:552-554.
18 Yazbek J, Raju SK, Ben-Nagi J, Holland TK, Hillaby K, Jurkovic D. Effect of quality of gynaecological ultrasonography on management of patients with suspected ovarian cancer: a randomised controlled trial. Lancet Oncol 2008;9:124-131.
Copyright/license for publication
The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, an exclusive licence on a worldwide basis to the BMJ Publishing Group Ltd and its licensees, to permit this article (if accepted) to be published in BMJ editions and any other BMJPG products and to exploit all subsidiary rights, as set out in BMJ's licence (http://resources.bmj.com/bmj/authors/checklists-forms/licence-for-public...).
Competing interest declaration
All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare no competing interests.
BVC is a postdoctoral fellow of the Research Foundation - Flanders (FWO)
Table 1. Recommendations related to ovarian cancer diagnosis from NICE clinical guideline 122 on recognition and initial management of ovarian cancer
First tests for diagnosis in primary care (p8)
? Measure serum CA125 in primary care in women with symptoms that suggest ovarian cancer.
? If serum CA125 is 35 IU/ml or greater, arrange an ultrasound scan of the abdomen and pelvis.
? If the ultrasound suggests ovarian cancer, refer the woman urgently for further investigation.
? For any woman who has normal serum CA125 (less than 35 IU/ml), or CA125 of 35 IU/ml or greater but a normal ultrasound:
o assess her carefully for other clinical causes of her symptoms and investigate if appropriate
o if no other clinical cause is apparent, advise her to return to her GP if her symptoms become more frequent and/or persistent.
Tumor markers to help establish the diagnosis in secondary care (p9)
? Measure serum CA125 in secondary care in all women with suspected ovarian cancer, if this has not already been done in primary care.
? In women under 40 with suspected ovarian cancer, measure levels of alpha fetoprotein (AFP) and beta human chorionic gonadotrophin (beta-hCG) as well as serum CA125, to identify women who may not have epithelial ovarian cancer.
Malignancy indices to help establish the diagnosis in secondary care (p9)
? Calculate a risk of malignancy index I (RMI I) score (after performing an ultrasound) and refer all women with an RMI I score of 250 or greater to a specialist multidisciplinary team.\
Research recommendation (p15)
? Further research should be undertaken to determine the optimum RMI I threshold that should be applied in secondary care to guide the management of women with suspected ovarian cancer.
Procedures for imaging in the diagnostic pathway (p9)
? Perform an ultrasound of the abdomen and pelvis as the first imaging test in secondary care for women with suspected ovarian cancer, if this has not already been done in primary care.
? If the ultrasound, serum CA125 and clinical status suggest ovarian cancer, perform a CT scan of the pelvis and abdomen to establish the extent of disease. Include the thorax if clinically indicated.
? Do not use MRI routinely for assessing women with suspected ovarian cancer.
Competing interests: No competing interests
The recent National Institute for Health and Clinical Excellence
(NICE) guidelines1,2 on the recognition and early detection of ovarian
cancer have generated considerable comment in the primary care and
gynaecological oncology communities, as evidenced by the letters to the
Editor last week3,4. Representing the organisations which practice,
promote and advance gynaecological oncology in the United Kingdom, we
welcome the development of guidelines to heighten awareness of the
challenges in making a diagnosis of ovarian cancer. We acknowledge the
limitation of evidence base on which the NICE guidelines were derived, and
appreciate the rationale for promoting access to CA125 in primary care as
a strategy to reduce the time to diagnosis. However, we believe it is
essential to promote awareness amongst our primary care colleagues of the
role and limitations of CA125 assessment in the detection and diagnosis of
ovarian cancer, and to clarify the context of the NICE recommendations
pertaining to CA125.
In order to improve the survival of women diagnosed with ovarian
cancer, we are keen to promote an understanding amongst women of the
symptoms which appear to be frequently associated with the onset of the
disease, in particular a frequency (more than 12 times per month) and
persistence of the following symptoms: abdominal distension / bloating,
early satiety / loss of appetite, pelvic or abdominal pain and urinary
urgency or frequency. When women with these symptoms present to primary
care, they should be examined and evidence of a pelvic mass (not thought
to be uterine fibroids) or ascites should trigger a "rapid access"
The CA125 test in isolation, promoted by the NICE guidance, is only
applicable to women with normal examination findings. This test will
identify some additional cases of ovarian malignancy where examination was
unremarkable, but the test is not specific for ovarian cancer and a range
of other diagnoses (benign or malignant) may be made following subsequent
investigations. Where the CA125 is raised (35 IU/ml or higher), an
abdominal and pelvic ultrasound should be performed within two weeks by an
appropriately trained and accredited sonographer.
Most importantly, as specialists in gynaecological cancer practice
we wish to stress that CA125 is normal in approximately one third of cases
of early stage ovarian cancer5. Thus a normal CA125 does not exclude
ovarian cancer in a symptomatic woman. The NICE guidelines indicate that
women should return to their GP if symptoms persist. Whilst neither
wishing to alarm women, or encourage excessive investigation of symptoms
that may not be due to ovarian cancer, we believe that the importance of
an early review by the GP is not sufficiently stressed in the NICE
guidelines. We propose that in symptomatic women a follow-up appointment
is arranged six weeks after a normal CA125 test, in order that symptoms
can be reassessed. Persistent symptoms at that time would indicate the
need for abdominal and pelvic ultrasound, and this should be done within 4
weeks; a negative scan result excludes ovarian cancer with a high degree
1. National Institute for Health and Clinical Excellence.
Recognition and initial management of ovarian cancer. CG122. 2011.
2. Redman C, Duffy S, Bromham N, Francis K; on behalf of the Guideline
Development Group. Recognition and initial management of ovarian cancer:
summary of NICE guidance. BMJ 2011;342:d2073.
3. Olaitan A. NICE on ovarian cancer. Recommendations for detection in
primary care are flawed. BMJ 2011; 342:d3022
4. Cave JA. NICE on ovarian cancer: Please include GPs in developing
5. Woolas RP, Xu FJ, Jacobs IJ, Yu YH, Daly L, Berchuck A, et al.
Elevation of multiple serum markers in patients with stage I ovarian
cancer. J Natl Cancer Inst 1993;85:1748-51.
Competing interests: No competing interests
Specialist commentary such as this will be welcomed by GPs who are
looking to avoid the rare but disastrous situation of missing an ovarian
However, I would caution Dr Olaitan about making sweeping
generalisations such as 'GPs have lost the skill of clinic pelvic
assessment'. A more realistic assessment is that GPs are unlikely to try
to examine every women presenting with the vague symptoms that are the
hallmark of ovarian cancer (and to find a chaperone to assist) when trying
to carry out a holistic consultation within 10mins.
Advice on how to improve on the NICE guidance (if you order a CA125,
order an USS) is welcome. I'm afraid that off-the-cuff generalisations are
Competing interests: No competing interests
I write in response to the letters, "Recommendations for detection in
primary care are flawed" and "Please include GPs in developing
guidelines", which were published in the BMJ on 21 May 2011 regarding
NICE's clinical guideline on the recognition and initial management of
Nearly 4,500 women die every year from ovarian cancer, in many cases
because their disease was not detected before it reached an advanced
stage. The guideline is intended to reduce the number of women dying
because they were not diagnosed in time.
The evidence shows that the CA125 blood test is the most reliable and
widely used tumour marker currently available. Given that the four key
symptoms of ovarian cancer can be attributed to a variety of conditions,
it would be incredibly costly and would bring little clinical benefit for
clinicians to arrange ultrasounds as a first port of call for every
patient who has experienced them. Most women who report having these
symptoms will not have the disease.
The key to suspecting ovarian cancer is in the persistence of the
symptoms. The guideline advises GPs to consider investigating women
(particularly those aged 50 and above) who report having any of the
symptoms on a persistent or frequent basis, i.e. more than 12 times a
If serum CA125 is 35 IU/ml or greater, an ultrasound of the abdomen
and pelvis should then be arranged. Women whose serum CA125 is less than
35 IU/ml should be carefully assessed for other clinical causes of the
symptoms, as appropriate. If no other clinical cause is apparent, the
patient should be advised to return if symptoms become more frequent
and/or persistent. It is impossible for any guidelines to cover every
clinical eventuality and so it is important for GPs to use their
professional judgement first and foremost when deciding how to treat or
refer their patients.
As with all NICE guidance, the recommendations have been based on the
best available evidence and have been developed by clinical and patient
experts. This included active input from a GP throughout the process. The
published guideline also followed a public consultation, in which anyone
with an interest in the topic was invited to comment on the initial
findings via one of the registered stakeholders. This included the Royal
College of General Practitioners and several other bodies where GPs would
have had the opportunity to provide their feedback. It is grossly
inaccurate to claim that the guideline has been developed without their
As Chair of the Guideline Development Group, I hope that the clinical
guideline will support GPs in their initial investigations so that women
with ovarian cancer can be referred to the appropriate specialist teams
sooner and begin treatment.
For further information about NICE's clinical guideline, including
the full list of stakeholders and how organisations can register, please
visit: www.nice.org.uk/CG122. This link also contains useful support tools
specifically created for GPs to help them implement the advice.
Mr Sean Duffy MBBCh, MD, FRCS, FRCOG
Chair of the Guideline Development Group responsible for developing
the recommendations on NICE's behalf and Medical Director of the Yorkshire
Competing interests: No competing interests