Shifting the goalposts on cancer diagnosis

NICE is urging GPs to send thousands more patients for investigations for potential cancer every year, in a bid to boost early diagnosis.

In revamped guidelines for suspected cancer, NICE says it expects GPs to consider cancer more readily in patients with undifferentiated symptoms and provides a raft of new recommendations on referrals and tests.

National media have jumped on NICE claims about the likely impact, with headlines declaring the guidelines will ‘save 5,000 lives a year’ and end the UK’s ‘shameful record’ on cancer survival.

But some GPs are warning of the risk of over-investigating people in whom cancer is unlikely and missing people who have a higher baseline risk.

Professor Willie Hamilton, a GP in Exeter and clinical lead for the NICE cancer guidelines, says the guidance is specifically aiming to eliminate the excess deaths in the UK attributed to diagnosis delays each year.

Professor Hamilton says: ‘It is generally well accepted that 10,000 lives are lost each year to cancer compared with the European average. About half of that figure is ascribed to diagnosis – so we have a target of 5,000 lives a year and this guidance is aimed directly at that.’

The guidelines form part of the Government’s drive to improve early cancer diagnosis, which includes pilots of patient self-referral for diagnostic tests and new multidisciplinary diagnostic centres. But the main target is to increase GP referrals for cancer.

As part of this, the NICE guidelines have been radically overhauled so they organise recommendations for GPs according to the symptoms their patients may present with in primary care – rather than by starting from the suspected cancer type and working back.

Lower threshold

Crucially, they also lower the threshold of risk at which GPs should refer patients to an urgent two-week cancer pathway – from symptoms and signs with a positive predictive value (PPV) of roughly 5% in the previous guidelines to a PPV of 3%.

And they now provide explicit ‘safety netting’ recommendations on when to order diagnostic tests in primary care for patients with symptoms that could signal cancer, but who fall below the 3% PPV.

This will help earlier diagnosis of bowel cancer in particular, according to Professor Hamilton, as GPs will be expected to order faecal occult blood testing (FOBT) in patients with between a 1% and 3% risk – so with symptoms like mild anaemia, abdominal pain or, for younger people, a change of bowel habit.

‘This guidance aims to save 5,000 lives a year’

Professor Willie Hamilton

Professor Hamilton says this group currently tends to slip through the ‘diagnostic net’.

He says: ‘The GP really struggled to get definitive investigations for this group of patients – FOBT steps into that gap. For me that is one of the biggest changes – and is one that will really matter because it is this group that often proceeds to emergency presentations.’

Other major changes include urgent access to MRI scans for patients with a loss of central neurological function and a new recommendation to send people with haemoptysis directly for a two-week urgent referral.

‘The risk from haemoptysis is high enough that it warrants really thorough investigation – so the specialist may wish to do a CT-scan or other tests over and above the chest X-ray.’

Dr Steve Hajioff, chair of the guidelines panel and a director of public health, says the new symptoms-based approach will make the guidelines ‘easier to use by primary care clinicians in a busy consulting room’.

But some GP experts are not convinced.

Professor Julia Hippisley-Cox, professor of epidemiology and general practice at the University of Nottingham and lead researcher for the QCancer risk algorithm, says the panel has not addressed ‘concerns about whether such long and complicated guidance could ever be implemented by GPs in everyday clinical practice’.

There are also concerns that the new approach does not take into account risk factors like family history or smoking.

Dr Nick Summerton, a former NICE clinical advisor and GPSI in cancer, who practices in East Yorkshire, says this approach is ‘bonkers’. He says: ‘To ignore baseline risk will mean increased referral of those at low risk and reduced referral of those at high risk – given the same symptoms. This is frankly stupid.’

Some experts fear changes to existing recommendations for specific cancer referral criteria will actually lead to both over- and under-referral.

Dr Jonathan Rees, a GPSI in urology in Nailsea in North Somerset and chair of the Primary Care Urology Society, says he is ‘very supportive’ of Professor Hamilton’s research on improving cancer diagnosis, but adds ‘there are changes to the guidelines that concern me’.

In particular, he says NICE has ignored urology experts’ concerns that restricting referral of patients with haematuria to older patients will lead to delays in bladder cancer diagnoses. Dr Rees also warns changes to PSA testing will leave men open to over-investigation.

NICE rejects these criticisms, insisting the guidance is based on the best available evidence from primary care.

Overstretched services

Whether GPs back the changes or not, the push to refer more low-risk patients means the number of people referred and tested through diagnostic services can be expected to increase significantly.

NICE has ruled out making a proper impact assessment because experts’ estimates on the change in referral and testing activity vary so widely.

However, Professor Hamilton says he believes there will be a ‘modest’ increase of ‘no more than half a million’ referrals over ‘a few years’, while the increase in direct access tests depends on what services are offered. 

Nevertheless, it is clear GPs in many areas currently have limited or no direct access to tests in primary care and GPs are already expressing concern at the impact on their local diagnostic services.

The main areas in which NICE expects to see increases are urgent lung cancer referrals – to accommodate new referrals of people over 40 with unexplained haemoptysis – as well as direct access tests for upper gastrointestinal tract cancers and bowel cancer.

A costing statement accompanying the guidelines acknowledges they will ‘increase demand on providers for additional capacity to undertake more diagnostic tests, on services that already have limited capacity’.

However, it says increases in resource use will likely be offset, for example, through reduced emergency admissions for lung cancer and streamlining of pancreatic cancer urgent referrals after an abnormal CT scan.

Dr Ishani Patel, a member of the Cancer Commissioning Support Team for London and a GP in north-west London, says the guidelines will help clarify what is required.

‘We will have to work with CCGs to deliver it,’ she says.