Professor Willie Hamilton outlines how revised recommendations from NICE can improve GP referrals for cancer testing and support earlier diagnosis
After nearly 3 years of deliberations, updated NICE guidance on selection of patients for cancer investigation has been published as NICE Guideline (NG) 12.1 The guidance is wide-ranging, covering 37 cancer sites, including all the most common adult ones; it also contains explicit recommendations for children. The decision to revise the previous NICE guidance (Clinical Guideline [CG] 27, published in 2005), was made for a number of reasons:
- Cancer survival figures in the UK remain stubbornly behind those in mainland Europe. Although UK metrics are improving, we have hardly narrowed the gap with other countries that have a strong system of primary care healthcare, except perhaps in breast cancer. Much of this has been ascribed to diagnostic differences between countries.2
- The original NICE clinical guideline (CG27; published in 2005) focused on referral as the main method of obtaining a cancer diagnosis, whereas in reality GPs (and other primary care professionals, including practice nurses) are the clinicians who suspect cancer: secondary care confirms (or refutes) it; furthermore, there are now primary care tests we commonly use (e.g. chest X-ray, prostate specific antigen, CA125, etc) and, although they are not definitive, a positive result takes us much closer to confirming the diagnosis.
- It has become increasingly clear that as initial identification of the possibility of cancer takes place largely in primary care—requiring primary care research to identify which features of cancer actually matter—there has been a dramatic increase in such research in the last decade, without which the updated NICE guidance would have been impossible.1
Organisation of recommendations by site and symptoms
There are two parallel methods of organisation of the recommendations in NG12. The first one is the structure we are familiar with from the 2005 guideline: recommendations are listed by cancer site, and then the various symptom profiles and actions are detailed. This is logical, as if a GP suspects a diagnosis of cancer, it is important to know which site may be involved, so investigation or referral can be directed to the correct department. However, patients present with symptoms, not with 'possible cancer', so the guideline also organises recommendations according to symptoms and findings from initial investigations, and include, the possible cancer sites and recommended actions. Organising recommendations by symptoms is also important to reduce the chance of patients with a particular symptom common to several cancers having sequential referrals to different departments (the so-called 'ping pong patient', particularly affecting those with pancreatic cancer).
As with all NICE guidance, the strength of the evidence underpinning each recommendation is indicated by the wording. Firm statements like 'refer' are used when there is sufficient evidence to support the recommendation, whereas the word 'consider' is used where the evidence is weaker or absent (see Strength of recommendations in the guideline).1 In practice, it is likely that this distinction will be immaterial once the guidance is implemented.
One final, and in my view, crucial point is that an explicit statement is made within the NICE guideline that the guidance does not override the clinician's individual responsibility to make decisions appropriate for the patient sitting in front of them.1 The 2005 guideline was rather regarded as Holy Writ—especially in policing entry into the 2-week clinics. This new NICE guideline accepts that exceptions will occur, and that the GP should be able to obtain an urgent opinion if, in their view, the patient needs one, without having to make the patient fit the recommendations.
It was not possible, or desirable, for NICE to cover the presentation of every one of the 330,000 new cancers diagnosed annually in the UK (the full version of NG12 plus appendices already exceeds 3000 pages).
Risk and investigation
The 2015 NICE guideline (NG12) uses a risk of cancer of 3% or more to underpin recommendations for urgent investigation or referral.1 There were very few recommendations in the 2005 guidance relating to risks below 5%, so this drop in threshold represents a liberalisation. To some extent, this has already happened: with awareness campaigns, risk-assessment tools, QCancer, and other initiatives being so prominent in recent years, the threshold for investigation has fallen, and times to diagnosis have fallen modestly too.3
The threshold of 3% was lowered for two other groupings:1
- the first being children, where very few presentations actually equate to a cancer risk of >3%, and where the rewards for successful cancer diagnosis and cure are much greater (children have longer to live)
- the second grouping was where good primary care tests are available and are inexpensive; thus it seemed sensible to liberalise recommendations for chest X-ray, prostate-specific antigen, and similar tests
- no specific risk threshold was proposed for primary care tests, though in practice few examples below 1% were recommended.'
Key updates to NICE recommendations
There are over 170 recommendations in NG12, so any selection of 'headline' ones has to be a personal one. I have concentrated on cancer sites where there have been major changes.
Possible lung cancer
Chest X-rays are recommended urgently (within 2 weeks) for those aged over 40 years, with two of the following symptoms if a non-smoker, and one if they are a smoker or ex-smoker:1
- shortness of breath
- chest pain
- weight loss
- appetite loss.
These are non-controversial, although clearly many smokers cough, so the GP will have to use their clinical judgement on when it is appropriate to investigate. This change is not that dramatic, as GPs have already increased the number of chest X-rays they have been requesting.4 However, there is a new recommendation for patients aged over 40 years and reporting unexplained haemoptysis to have a 2-week respiratory referral for lung cancer.1 The summary positive predictive value was over 3%, justifying this.5 NICE did not suggest which tests the specialist may do, but it is possible that specialists may choose to proceed to a computed tomography (CT) scan if the chest X-ray is negative.
Possible colorectal cancer
There are a few more recommendations for 2-week referral for patients with suspected colorectal cancer—most of whom will receive a colonoscopy (although, again, NICE does not make suggestions for investigation once referral has been recommended). NICE NG12 recommends 2-week referral for patients:1
- aged 40 years and over with unexplained weight loss and abdominal pain
- aged 50 years and over with unexplained rectal bleeding (note the removal of the previous recommendation in CG27 for the patient to have had 6 weeks of bleeding, but the addition in NG12 of 'unexplained')
- aged 60 years and over with iron-deficiency anaemia or changes in their bowel habit (no threshold haemoglobin value is now quoted, with the thresholds in CG27 generally accepted as representing too severe anaemia6).
Some groups of patients were ill-served in the previous NICE guideline (CG27)—particularly patients under 60 and those with abdominal pain or mild anaemia. They did not qualify for colonoscopy, had longer times to diagnosis,3 were more likely to have an emergency presentation,7 and consequently had higher mortality.8 The risk of cancer in these patients is low—below 3%, but not zero. Five studies of faecal occult blood testing in these groups (who do not qualify for colonoscopy) supported its use, so this is now recommended for patients:1
- aged 50 years and over with unexplained abdominal pain or weight loss
- aged under 60 years with changes in their bowel habit or with iron-deficiency anaemia
- aged 60 years and over with anaemia even in the absence of iron deficiency.
The wording does not specify what test is to be used for occult blood, leaving open the possibility of switching to faecal immunochemical tests should these prove to be superior in this symptomatic population.
Possible oesophageal cancer
The main update in the 2015 NICE guideline is to offer urgent direct access gastroscopy (to be performed within 2 weeks); this compares with the 2-week referral recommended in the 2005 guideline, which almost always led to a gastroscopy anyway. Assessment is recommended for those patients:1
- with dysphagia (this remains the main indication for assessment)
- aged 55 years and over with weight loss accompanied by upper abdominal pain, dyspepsia, or reflux.
Non-urgent endoscopy to assess for oesophageal cancer is recommended for several combinations of upper gastrointestinal symptoms in those aged over 55 years, including treatmentresistant dyspepsia, and for people with haematemesis.1
Other notable new recommendations
Finally, NICE NG12 recommends open-access urgent (within 2 weeks) magnetic resonance imaging (MRI) of the brain (or CT if MRI is contraindicated) to assess for brain or central nervous system cancer in adults with progressive, sub-acute loss of central neurological function.
The new guideline also suggests consideration of urgent (within 2 weeks) direct access CT scan (or ultrasound if CT is unavailable) for possible pancreatic cancer in those aged 60 years and over with weight loss and a second relevant symptom.1
Both of these new recommendations will require organisational change, but as 2-week-wait clinic appointments are side-stepped, it should be possible to perform these tests more cheaply. It is also patient-friendly as the process is simpler for the patient.
The 2015 NICE guideline NG12 will take some time to implement, although few of the changes are revolutionary. The hope is that implementation of the recommendations from NG12 will further support improvement in cancer outcomes in the UK, which will match or exceed the rest of Europe.
- Cancer survival figures in the UK remain stubbornly behind those in mainland Europe2
- Differences in diagnosis may be the reason behind the poorer UK performance2
- Referrals for testing from primary care healthcare professionals can help to reduce the time to diagnosis
- The 2015 NICE guideline NG12:1
- organises recommendations by site and symptoms; the latter may help to reduce the chance of patients with a particular symptom common to several cancers having sequential referrals to different departments
- highlights that the guidance does not override the clinician's individual responsibility to make decisions appropriate for the patient sitting in front of them—the GP should be able to obtain an urgent opinion if, in their view, the patient needs one
- uses a risk of cancer of 3% or more to underpin recommendations for urgent investigation or referral—this compares with very few recommendations in the 2005 guidance related to risks below 5%
- Some new recommendations will require organisational change, but in cases where the 2-week wait clinic appointments are side-stepped, it should be possible to perform tests more cheaply
- Moving straight to testing without first requiring a secondary care referral is also patient-friendly as the process is simpler for the patient.
GP commissioning messages
written by Dr David Jenner, NHS Alliance GMS contract/PBC Lead
- NICE Guideline 20 will require commissioners to:
- review all current cancer referral pathways
- ensure direct access pathways, to which GPs can make direct referrals, are commissioned (in particular for MRI, CT, and endoscopy)
- Less referrals to some of the 2-week-wait suspected cancer pathways can be anticipated (e.g haematuria) but in the majority of instances, referral volumes are likely to increase
- With pressure on many of the key nationally defined performance targets related to cancer (i.e. diagnostics, 2-week wait, and 31- and 62-day referral targets), commissioners and providers will need to:
- scope out and respond to the potentially increased demand
- ensure there is no further deterioration in performance against these targets
- Production and use of a patient information sheet that explains the importance of prompt investigation for suspected cancer could be developed to help avoid missed appointments, particularly as targets do not allow for such exceptions
- Educational programmes should be made available for GPs and other primary care professionals to increase understanding of the new guidance and subsequent commissioned suspected cancer referral pathways.
MRI=magnetic resonance imaging; CT=computed tomography
- NICE. Suspected cancer: recognition and referral. NICE Guideline 12. NICE, 2015. Available at: nice.org.uk/guidance/ng12↩
- Coleman M, Forman D, Bryant H et al and the ICBP Module 1 Working Group. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995–2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. Lancet 2011; 377 (9760): 127–138. ↩
- Neal R, Din N, Hamilton W et al. Comparison of cancer diagnostic intervals before and after implementation of NICE guidelines: analysis of data from the UK General Practice Research Database. Br J Cancer 2014; 110 (3): 584–592. ↩
- NHS England. Factsheet: Early diagnosis of cancer by delivering improved access to diagnostics. NHS England, 2014. Available at: www.england.nhs.uk/wp-content/ uploads/2014/02/sm-fs-4-41.pdf. ↩
- National Collaborating Centre for Cancer.Suspected cancer: recognition and referral. NICE Guideline 12. NICE, 2015. Available at: www. nice.org.uk/guidance/NG12/evidence↩
- Hamilton W, Lancashire R, Sharp D et al. The importance of anaemia in diagnosing colorectal cancer: a case-control study using electronic primary care records. Br J Cancer 2008; 98: 323–327. ↩
- Cleary J, Peters T, Sharp D, Hamilton W. Clinical features of colorectal cancer before emergency presentation: a population-based case-control study.Fam Pract 2007; 24: 3–6. ↩
- Stapley S, Peters T, Sharp D, Hamilton W. The mortality of colorectal cancer in relation to the initial symptom at presentation to primary care and to the duration of symptoms: a cohort study using medical records. Br J Cancer 2006;95: 1321–1325. ↩