Progress across the cancer pathway from 2000

This section describes the main developments since 2000, across the main dimensions described in the various cancer strategies, from prevention, screening, diagnosis and treatment, to living with and beyond cancer. One deliberate omission is end of life care, which merits its own analysis. Each section draws on available data, including the testimony of those people directly involved in developing and implementing policy.

3-1 Prevention

Around 4 out of 10 cancers are caused by exposure to risk factors, such as smoking, that are preventable. The scope of prevention has broadened considerably from the NHS Cancer Plan, which focused heavily on reducing smoking and increasing fruit and vegetable consumption, with only brief mentions of the role of alcohol, obesity, lack of exercise and exposure to the sun as cancer risks. The most recent cancer strategy (2015) has been developed on the basis of a much improved evidence base, particularly in relation to the effects of obesity and alcohol consumption on cancer, as well as action to reduce exposure to ultraviolet radiation and boost the coverage of the human papilloma virus (HPV) vaccination. The next section looks in more detail at the action taken on two of the main risk factors: smoking and being overweight or obese.

3-1-1 Smoking

Action to reduce smoking rates represents one of the most successful strands of health policy in recent years. It is the product of successful mobilisation of a broad set of actors, which included the cancer community, but extended beyond to a coalition of health charities, politicians (able to vote freely) and the research community.

If I look back at my time involved with the cancer community, it [smoking] stands out as the biggest success story, that we’ve changed the nature of the debate, and we have got politicians to do things we never thought they’d do.

Sarah Woolnough, Executive Director of Policy and Information, Cancer Research UK

Smoking rates are now at their lowest levels since records began (Figure 15). In 2016, 18% of adults in England smoked cigarettes (20% of men, 16% of women), compared with 28% in 1998 when Smoking kills: a white paper on tobacco was published.

For the UK this represents significant progress. Following the Second World War the UK had a particularly high prevalence of cigarette smoking in men: more than 60% in 1948. In 1950 a seminal paper, Smoking and carcinoma of the lung, was published in the BMJ, linking deaths from lung cancer with smoking tobacco. As further proof of the link emerged, smoking rates in men began to fall, and in 1965 the advertising of tobacco on TV was banned.

Figure 15: Changes in cigarette smoking prevalence in male and female adults (16 and older) between 1993 and 2016

Source: Data provided by Cancer Research UK. Source: Health Survey for England; NHS Digital. Available at:

For almost 40 years after the first evidence of harm from smoking appeared, central government made relatively little effort to regulate tobacco. Taxes on tobacco crept up slowly and, other than the ban on television advertising, the industry was relatively unrestricted. By the mid-1990s, it was clear that the prevalence curves had flattened, and smoking was no longer declining.

In 1997 the Labour Party manifesto included a pledge to ban tobacco advertising, and the first-ever tobacco control strategy, Smoking Kills, was published in December 1998. In the following 10 years, action was taken on smuggled cigarettes, stop smoking programmes, public awareness programmes, smoke-free legislation (2007), a ban on smoking in public places, compulsory health warnings on cigarette packets, standardised packaging and a ban on advertising in shops.

Looking back, for one of the leading civil servants, the period represents a positive feedback loop between evidence and action:

If you look at 2002–2007 there was an acceleration [in tobacco control], as more and more of the properly evidence-based policies happened. There was an acceleration, but I could not at all have guessed that it would end up at the rate of reduction. It’s the smoke-free legislation. Part of that, and the whole denormalisation piece, is about creating a societal movement that it [smoking] is not the right thing, or normal thing, to do and that’s what you hope for.

Nick Adkin, former Head of Tobacco Programme, Department of Health

An important part of the NHS Cancer Plan was establishing smoking cessation services in the NHS, reversing the ban on prescribing nicotine replacement therapy, and putting in place financial incentives for GPs to register smokers and refer people to cessation services. According to the latest data, numbers accessing NHS stop-smoking services rose steadily until 2011/12, but have since declined. It is not clear what is driving this. Since 2012, public health has fallen in the remit of local authorities, which have seen their total budgets fall by 32.6% between 2011/12 and 2016/17. While some of the drop-off in smoking cessation numbers may represent under-reporting from local authorities, NHS Digital estimates that access to e-cigarettes has had a role to play. So too might underinvestment in the stop-smoking services previously supported by local government – between 2014/15 and 2018/19 there has been a 32% reduction in real term spend on stop-smoking services.

Figure 16: Number of smoking quit attempts (characterised as setting a quit date) and the percentage of those that were self-reported as being successful between 2007/08 and 2017/18, England

Note: The 2016/17 and 2017/18 data have not been adjusted to estimate for the local authorities that did not provide any data or only provided data for some quarters. These totals are therefore underestimates and not directly comparable with previous years.

Source: Lifestyle statistics; NHS Digital. Available at: (accessed on 19 October 2018).

There was also a significant investment in media campaigns from 2003. Charities, including Cancer Research UK and the British Heart Foundation, funded advertisements which produced memorable images such as children breathing in smoke, and cigarettes dripping fat.

By 2004, the government was being confronted with an important lesson, which was that voluntary agreements with business and industry had only limited traction, in this case to provide smoke-free areas in places of work, which was part of the 1998 Smoking Kills strategy. What seemed to have helped shift this was the impact of the media campaigns on public attitudes, as evidence mounted from Office for National Statistics surveys that the public welcomed smoke-free areas. As Nick Adkin recalls, ‘what we were trying to do is generate the evidence base for ministers to be more comfortable with a decision which was about legislating in this area.’

When Ireland and Scotland decided to ban smoking in public places, it tipped the weight of evidence even further. The impact of Ireland is crucial in this.

So, Irish ministers. Huge problems with smoking, and drinking [in Ireland]. They made a decision to go for, ‘This is a flagship policy for Ireland as modern Ireland. You know, forward thinking, not backward-looking smoky pubs. This is our future.’

Nick Adkin, former Head of Tobacco Programme, Department of Health

Behind the symbolism of Ireland’s decision came evidence, funded by Cancer Research UK.

I remember there were big concerns about the impact on the economy and on pubs, of the workplace ban. So, we funded an evaluation of the Irish Republic who’d gone earlier, and that was often quoted. Politicians took big interest in that, to be able to say, ‘Look. The pub trade will survive this,’ was incredibly important, politically.

Professor Sir Alex Markham, Professor of Medicine, University of Leeds, and former Chief Executive of Cancer Research UK

The first explicit call for legislation came in the Chief Medical Officer’s report of 2004. The legislation was eventually passed with a majority of 200 in 2006. Rapid progress was then made on the next anti-smoking goals, such as point-of-sale restrictions and plain packaging, as stakeholders learned how to marshal arguments effectively:

Actually, when it [point-of-sale restrictions] came to it, it certainly wasn’t as hard a sell as the workplace smoking ban. I think partly because the community – and it was a very broad community – had done such a good job in using the evidence and what we knew about public opinion, to try to translate that into a strong story to go to government with.

Sarah Woolnough, Executive Director of Policy and Information, Cancer Research UK

Smoking continues to be the leading cause of preventable deaths, and currently causes around 50,000 cases of cancer per year in England (across 15 cancer types). Men are more likely to smoke than women, younger people more likely to smoke than older people, and smoking rates are significantly higher in more deprived communities and among people with mental illness.

The most recent tobacco control plan for England, Towards a smokefree generation, published by Public Health England in July 2017 pledges that, by the end of 2022, it will:

  • reduce the prevalence of 15-year-olds who regularly smoke from 8% to 3% or less
  • reduce smoking prevalence among adults in England from 15.5% to 12% or less
  • reduce the inequality gap in smoking prevalence between people in routine and manual occupations and the general population
  • reduce the prevalence of smoking in pregnancy from 10.7% to 6% or less.

3-1-2 Obesity

Being overweight or obese has now been recognised as the next biggest preventable cause of cancer behind smoking. Researchers estimate that obesity causes around 23,000 cancer cases each year in the UK, and being overweight or obese is linked to at least 13 different types of cancer, including bowel and endometrial cancers., Despite this, public awareness of the relationship between obesity and cancer remains low, with just 15% of UK adults aware of the causal link.

In contrast to tobacco, the evidence base is much newer. Obesity merited only a single bullet point in the NHS Cancer Plan. In the 2015 cancer strategy it is given much more prominence. The learning from the smoking campaigns has been vital. Similar tactics to those deployed in tobacco control are being developed by government agencies, such as Public Health England, the Royal Colleges and charities, but there is still much ground to cover.

It’s not rocket science. If you go after the availability of high fat, sugar, salt foods and try to do something about that, you tackle the price, you tackle the promotion, you know, it’s the four Ps (product, price, place and promotion) that we’ve always referred to in tobacco control, you will make a difference. We do have a job to do, to change the environment in the way that we’ve done for tobacco, with food.

Sarah Woolnough, Executive Director of Policy and Information, Cancer Research UK

Weight is a growing problem in the UK. The Health Survey for England measures a representative sample of adults aged 16 and older to provide estimates of obesity levels in the country. The 2016 survey found that 26% of adults in England are obese (body mass index (BMI) of at least 30) and a further 35% are overweight (BMI of 25 to 29.9), meaning that 61% of people are either overweight or obese. As shown in Figure 17, the percentage of adults in England who are obese or obese and overweight combined has increased steadily since 1993. Cancer Research UK and the UK Health Forum have predicted that, if current trends continue, obesity will cause an additional 670,000 cases of cancer in the UK over the next 20 years.

In 2016 the government launched its childhood obesity strategy with the aim of ‘significantly reducing’ childhood obesity within 10 years. This included launching a ‘soft drinks industry levy’ – a levy on manufacturers of sugary drinks, implemented in April 2018. In February 2018 Cancer Research UK launched a national campaign aimed at raising awareness of the link between obesity and cancer. There is currently no national strategy for adult obesity.

Figure 17: Changes in prevalence of adults (16 and older) being obese and overweight between 1993 and 2016

* Overweight = BMI 25 to less than 30kg/m2. ** Obese = BMI 30kg/m2 or more.

Note: Data up to and including 2002 are unweighted; from 2003 onwards data have been weighted for non-response.

Source: Health Survey for England; NHS Digital. Available at: (accessed on 22 October 2018).

3-2 Screening

The aim of screening is to reduce mortality from cancer. This can be achieved either by detecting and removing precancerous lesions – thereby preventing cancer from developing – or by detecting cancers early, making a cure more likely than if they were detected once symptoms have appeared. For some cancers, the challenge is to balance the risk of overdiagnosis (finding cancers that wouldn’t cause problems in a patient’s lifetime and exposing them to potentially unnecessary treatment) against the benefits of earlier intervention. Evidence-based recommendations are made by a National Screening Committee, which is UK-wide. Screening programmes are nationally coordinated but locally commissioned, and are almost exclusively delivered by NHS trusts and GPs.

National screening programmes are currently available in the UK for three cancer types – breast, cervical and bowel cancers. The breast and cervical screening programmes were already well established by the time of the NHS Cancer Plan, and represented the only nationally coordinated aspect of cancer-related services before 2000.

It is important to recognise that only 29% of breast cancers and 10% of bowel cancers are currently detected by screening, and that uptake of screening remains far below 100% across all programmes. Across all cancers combined, only 6% of cases are detected through screening.

3-2-1 Breast cancer screening

The NHS breast screening programme was introduced in 1988. Women were invited for screening at 3-yearly intervals from around the age of 50 to around the age of 65, meaning that an individual woman would be invited five times in total. The NHS Cancer Plan extended the programme up to the age of 70 (increasing the number of times a woman is screened from five to seven), and promised a change from single to two-view mammography at all screening attendances by 2003. During this period the National Cancer Screening programme was responsible for the rollout of new screening services with considerable support and direction from the National Cancer Director and Department of Health Cancer Policy Team.

I was involved, hands dirty, you know, ringing up clinicians and hospitals, and getting into clinics. I was really into the details.

Julietta Patnick, former Director of National Screening Programmes

These changes led to a marked increase in cancers detected both at first and subsequent rounds, and had significant implications for the workforce. The NAO estimated it had resulted in a 40% increase in workload. Expanding the breast radiology workforce was not possible in the time frame. Instead the National Cancer Director and the Director of the Cancer Screening programme redesigned the workforce with the Department of Health and professional bodies, including introducing a four-tier model for radiographers – assistant practitioner, radiographer, advanced practitioner and consultant radiographer. This meant that trained radiographers could perform tasks previously done by radiologists.

Figure 18: Changes observed between 2006/07 and 2016/17 in the number of women (50–70) invited for breast screening, the number of women (50–70) screened and the percentage uptake

Source: Breast Screening Programme; NHS Digital. Available at: (accessed on 22 October 2018).

Figure 18 shows that, between 2006/07 and 2016/17, the total number of women being screened for breast cancer (of all ages) has risen by more than half a million because of the age extensions, even though uptake has fallen slightly. The UK performs above the Organisation for Economic Co-operation and Development (OECD) average for the number of women aged 50–69 who have received breast screening in the past 2 years though remains below nine countries, including Denmark, Norway and Sweden (Figure 19).

Further extension of the breast screening programme followed the publication of the Cancer Reform Strategy in 2007. The aim was to extend screening to a total of nine rounds from around age 47 to around age 73. As there was uncertainty about the benefits of screening in these younger and older age groups, it was decided that the extension should be introduced through a randomised controlled trial (AgeX). This has now recruited more than 3 million women. Full results of impact on mortality are not expected for several years.

Figure 19: International comparison of the percentage of women (aged 50–69) screened for breast cancer in the past 2 years (in 2015 or nearest year)

* The OECD average is calculated for the 33 OECD countries for which data were available.

Source: OECD Health Statistics (2017), EHIS Eurostat database. Available at: (accessed on 22 October 2018).

3-2-2 Cervical cancer screening

Although the Pap smear to screen for cervical cancers was developed in the 1940s, a national programme to invite women to attend their GP for screening at set intervals was only introduced in 1988. Since 2003, women aged between 25 and 49 receive screening invitations every 3 years, and those aged 50 to 64 every 5 years.

In 2000, the NHS Cancer Plan committed to change the technology used for cervical screening from smears to liquid-based cytology. This required retraining GPs and others to take samples and training of laboratory technicians (cyto-screeners) to interpret the samples. The programme was rolled out over several years.

Increased attention to the quality of screening programmes was galvanised early in the Labour government’s new term by the emergence of breast screening failures at a hospital in Exeter, which followed hard on the heels of a cervical screening scandal at Kent and Canterbury Hospital. The director of the screening programmes at the time recalled a swift response, which culminated in a strengthening of central power over the quality of screening happening in NHS trusts.

The vast majority of cervical cancers are caused by HPV, and in 2003 research on screening for HPV, either to replace cytology or as an adjunct to it, was commissioned. In England, samples from women with borderline or mild changes in the cells at the cervix now go on to have an HPV DNA test. Women who test positive for high-risk types of HPV are referred onward for colposcopy (an examination of the cervix where abnormal cells can also be removed). Those who are HPV negative are returned to routine screening.

Figure 20: Changes observed between 2006/07 and 2016/17 in the number of women invited for cervical screening (25–64), the number of women screened (25–64) and the percentage uptake

Note: Uptake was calculated by the Health Foundation by dividing the number of women screened by the number invited.

Source: Cervical Screening Programme. Available at: (accessed on 22 October 2018).

Since September 2008, girls aged 12–13 have been offered vaccination against HPV, and in 2016/17, 83.1% of eligible females completed the two-dose HPV vaccination course. In time, this should reduce the incidence of cervical cancer. Earlier this year, the government announced that the HPV vaccination will be rolled out to boys aged 12–13. There is ongoing research into HPV testing as a means of screening for cervical cancer (as opposed to its present use as an adjunct to the smear/cytology test).

Figure 20 shows that, between 2006/07 and 2016/17, the number of women screened for cervical cancer each year has fluctuated, and uptake has generally reduced over time. The increase in women screened in 2008/09 has been associated with the diagnosis and death from cervical cancer of the reality TV star Jade Goody. An increased screening attendance was observed at all ages but particularly for women under 50. The increased number of women invited for screening in 2011/12 may, in part, be explained by the increased number of women screened in 2008/09, who would have been expected to receive their next routine invitation for screening in 2011/12.,

Without the cervical screening programme, incidence of and deaths from cervical cancer would almost certainly have increased over this period as the incidence of the sexually transmitted HPV viruses has risen. It is estimated that, in England, screening currently prevents 70% of cervical cancer deaths. If everyone attended screening regularly, 83% of deaths could be prevented.,

3-2-3 Bowel cancer screening

Bowel (colorectal) cancer is the third most common cancer in both men and women in the UK, with an estimated 41,700 new cases per year. Cancer Research UK has estimated that 54% of bowel cancers are preventable, and screening is vital.

By the year 2000, pilot programmes were underway to test the use of faecal occult blood testing (FOBT) as screening for bowel cancer. These followed the publication of three trials all showing that bowel screening reduced mortality from bowel cancer. The evidence was sufficiently robust by 2006 for a national bowel cancer screening programme to be rolled out. FOBT kits are sent to people aged 60–69 every 2 years, with those testing positive for blood in the stool invited for a colonoscopy (diagnostic camera test). Following the Cancer Reform Strategy (2007) this was expanded to people aged 60–74. Recent Bowel Cancer Screening Programme data show an uptake rate of 59% for FOBT screening, with 1.66% of those participating testing positive for blood in the stool. In 2016/17, 3,021 people were diagnosed with a bowel cancer following screening, and a further 16,356 had potentially pre-cancerous lesions found.

In 2010 the results of a UK-based trial of an alternative approach to bowel screening were published, showing a reduction in mortality from colorectal cancer for patients aged 55 undergoing a one-off flexible sigmoidoscopy (camera test). With publication timed to coincide with the run-up to a general election, the introduction of flexible sigmoidoscopy became a manifesto commitment for both Labour and the Conservatives. Rollout subject to successful piloting was agreed, and this is now being delivered in around 65 centres. This has been harder than anticipated, in part due to a lack of endoscopy capacity, and in part as a result of poor uptake from the public.

The bowel screening programme is now undergoing further change, with FOBT being replaced by faecal immunochemical testing (FIT). This is an easier procedure from a patient’s perspective (it requires only a single stool sample, unlike FOBT which requires six samples from three separate stools) and has been shown to have a higher uptake than FOBT. Full rollout of FIT was due by April 2018 but has been delayed. The bowel cancer screening programme places demands both on colonoscopy and pathology services, as polyps removed at colonoscopy have to be analysed by histopathologists. The lack of workforce capacity and delays in procuring the FIT tests, is a source of frustration for the bowel cancer charities:

It [introduction of FIT] would result in a 120% increase in pathology demand in some areas, and we just don’t have the capacity. Now, we’ve seen this coming for years, but action hasn’t been taken until the last minute to prepare. Therefore, what they’re beginning to propose is that they will now, instead of doing a switchover from FOBT to FIT as there has been in Scotland, do a phased rollout as capacity becomes available. That could be 2 to 3 years for some areas, it’s such a lost opportunity.

Deborah Alsina, Chief Executive, Bowel Cancer UK

3-3 Earlier and faster diagnosis

By the late 1990s, some researchers and clinicians argued that survival rates in the UK were relatively poor compared with other countries due to delays in getting patients into treatment. Two systematic reviews were published in The Lancet in 1999 which suggested that delays in presentation with breast cancer were linked to poorer outcomes., But this was far from convincing for most cancer clinicians at the time, as one of the researchers recalls:

I suppose my overwhelming sense at the beginning of it all was the huge resistance of the cancer clinical community to this idea that delayed presentation or delayed diagnosis could in any way count for significant differences in outcome. The vituperative responses that our systematic review drew – the arguments were worse than politics or religion. These were core beliefs that were being challenged. ‘How could a 3-month delay possibly explain a reduced survival probability?’ So, it’s a wonderful example of evidence that, on its own, doesn’t win the day.

Amanda Ramirez, Professor of Liaison Psychiatry, King’s College, London

This resistance was difficult to overcome because of gaps in the data. Although cancer, compared with other diseases, had comparatively advanced datasets (having kept registries for several decades), in 1999/2000 it was not possible to know where the delays were happening (ie patients not going to the GP in the first place, delays between the GP and the hospital consultant, or delays in diagnosis once in the hospital system). In the event, because data on patients in the hospital system were easier to collect, much of the early effort from 2000 went into reducing delays in the hospital system (described in section 3.3.5). The puzzle of delayed diagnosis had to wait several years before new streams of research bore fruit. This work, some of it drawing on new comparative datasets, has since revealed multiple dimensions to the challenge of late diagnosis, and a problem still far from being solved.

3-3-1 Public awareness of symptoms and accessing GPs

Public awareness of which symptoms ought to be checked out by a GP is a self-evidently vital part of an effective cancer system. But in 2000, there was only a brief mention of the need for better public awareness in the NHS Cancer Plan. However, there was a commitment to produce a better evidence base.

By 2007, evidence from the EUROCARE studies, UK-based researchers and the Cancer Patient Experience Survey all pointed to patients being diagnosed later in England (and the UK) than in other countries. In response, the Cancer Reform Strategy established a multi-pronged National Awareness and Early Diagnosis Initiative, formally launched in 2008. The initiative had seven workstreams, including measuring public awareness and promoting interventions to prompt patients to go to their GP.

The measurement work resulted in the validation and first successful use of a national survey to test public awareness of cancer symptoms. The paper, published in 2009, found that overall cancer awareness was low, but that 68% of people could recall (unprompted) that a lump or swelling was a cancer symptom. The results also suggested that there was a problem right at the start of the cancer ‘pathway’ in getting people to go to their GP at all. In 2009, 38% of people were worried about wasting the doctor’s time, and 41% said it might be difficult to make an appointment. Cancer Research UK now publishes the Cancer Awareness Measure survey. In 2017, 45% of respondents said they found it difficult to make an appointment, 22% were worried about wasting the doctor’s time, and awareness of some cancer symptoms appears to have worsened: 59% of people recalled that a lump might be a cancer symptom.

These studies have also revealed systematic differences in knowledge according to age, gender, socioeconomic status and ethnicity: older, wealthier, white women are the most likely to know the risk factors, and seek help promptly. But the work also incorporated an international dimension, which has revealed factors that might be unique to the UK. Research conducted by the International Cancer Benchmarking Partnership revealed that low public awareness was a problem in Norway, Sweden, Denmark, Australia and Canada. But ‘not wanting to bother the doctor’ turned out to be a peculiarly British trait.

Since 2011, raising public awareness has been tackled through the ‘Be Clear on Cancer’ campaigns, run in collaboration with Public Health England, the Department of Health and Cancer Research UK. Time-limited national campaigns have been run on lung cancer, bowel cancer, bladder cancer and breast cancer (aimed particularly at older women above the age for routine breast screening). A wide range of resources, from leaflets to videos, have been developed for use by the NHS and other public services.

For cancer experts such as Professor Mick Peake, a lung cancer specialist, the campaigns have helped GPs as well as patients to navigate what has traditionally been a fraught relationship between primary and secondary care:

They [GPs] feel that the public awareness campaigns have given them the ‘permission’ to refer, at a lower threshold than they had before. You have seen hospital correspondence and know what these letters were like. The GP gets two letters, in the first the consultant says: ‘You’ve been sitting on this patient for months now and he’s got advanced disease, he’s going to die, you know, what have you been doing?’ Next day another letter arrives saying essentially: ‘What are you doing sending me this patient? There’s nothing wrong with him, what a waste of my time.’ You know, you can’t win.

Mick Peake, Professor of Respiratory Medicine and former Cancer Services Collaborative Lung Cancer Lead

3-3-2 Speeding up the journey from general practice onwards

In the absence of evidence about the precise nature and location of the barriers to early diagnosis, the early efforts of the NHS Cancer Plan focused instead on what happened after referral. This meant making sure that patients who did go and see their GP and were referred for a suspected cancer could be seen promptly. The NHS Cancer Plan took action based on existing evidence that there were unacceptable waits in three parts of the cancer pathway:

  • between referral and first hospital appointment
  • between first hospital appointment and key diagnostic test
  • between diagnosis and treatment.

This evidence was based on an audit of waiting times (run in 1997) and results from a pilot Cancer Patient Experience Survey (1999/2000). The audit, (published in 2000), looked at the experience of more than 13,000 patients referred by their GPs for cancer in October 1997. GPs already categorised patients into ‘urgent’ and ‘non-urgent’. On average, urgent cases were first seen by a specialist within 2 weeks, but people often faced additional long waits for treatment. Those patients referred as ‘non-urgent’ by their GP faced much longer waits, for first outpatient appointment and subsequent treatment. The audit also found large variations by tumour type and hospital trust.

The ‘2-week wait’ pledge was brought in to standardise the ‘urgent’ GP referral route, effectively forcing hospital consultants to accept the verdict of the GP that this was an urgent case. The first standard, for women with suspected breast cancer to be seen by a specialist within 14 days of urgent referral by their GP, was implemented in 1999. For other cancers, the same standard was applied in April 2000. Initially, around 600,000 patients were referred each year through this route.

Assessment of performance against this target was (deliberately) simple: 93% of people classified as urgent had to be seen within the required time frame. There was no specific intention to monitor the proportion of suspected cancer cases classified as urgent, or the proportion of cancers eventually found.

This target was very quickly met. By 2003, when progress against the NHS Cancer Plan was first reported, 98.5% of people with suspected cancer were being seen within 2 weeks, and subsequent performance has been consistently above the 93% standard (Figure 21), until the first quarter of 2018/19 when performance slipped for the first time in a decade (although this may relate to a rapid increase in urgent referrals of patients with possible prostate cancer, following media publicity around two celebrities (see 4.11) and possible breast cancer, following a national breast cancer campaign for women over 70 years of age).

Figure 21: Urgent GP referrals and proportion seen within 14 days between Q4 2008/09 and Q2 2018/19, England

Data provided by Cancer Research UK. Source: Cancer Waiting Times – National Time Series – Provider based; NHS England (2018). Available at: (accessed on 23 October 2018).

Figure 21 also shows a large increase in the number of patients being referred through the urgent, ‘2-week wait’ route. By 2017, almost as many people were being referred every 3 months as had been each year in the early 2000s. Now, almost 2 million go through this route per annum. Of these people, around 8% are found to have cancer.

3-3-3 Influencing GP referral rates

While an increase in the volume of urgent referrals was not an explicit aim of the 2-week standard, it was accompanied by newly developed guidance for GPs. By common consent, among those interviewed for this report, primary care was relatively neglected in the early cancer strategies, including Calman–Hine and the NHS Cancer Plan.

In these early years, I don’t think the penny had dropped that, unless we could tackle the impact of primary care on access to cancer diagnosis and manage the gatekeeping function of GPs more effectively, we wouldn’t be able to get the improvement in outcomes that we were seeking.

Peter Selby, Professor of Cancer Medicine, University of Leeds

This was partly a function of poor data on how patients were actually moving through the system, and a lack of appropriate guidance for GPs on how to interpret and spot symptoms. On average a GP will see around eight new cases of cancer each year, but will also see hundreds of patients with symptoms that could possibly be due to cancer.

The patients we send up the 2-week wait are the easy ones. We can do that with our eyes shut. Have they got red flag symptoms? Yes. The really difficult ones are the grey areas, the ones with only a 1–2% risk of cancer.

Dr Fiona Walter, Reader in Primary Care Cancer Research, University of Cambridge

The referral guidelines for suspected cancer were published in 2000 by the Department of Health, and covered all cancer types. These were sent to all GP practices and the breast cancer targets were the first to be achieved. However, these guidelines were largely based on the symptom profiles of patients diagnosed with cancer (in secondary care) not on those presenting in primary care.

In 2015 NICE published new guidelines (NG12) to help GPs know which patients should be referred urgently for suspected cancer. These reflect the growing body of primary care research, stimulated by the NHS Cancer Plan and National Awareness and Early Diagnosis Initiative, into the risk of cancer associated with individual symptoms or combinations of symptoms. The guideline sets a risk threshold of 3% as being appropriate for an urgent referral, though a recent large-scale study in primary care Prostate Cancer Intervention Versus Observation Trial (PIVOT) indicates that patients believe that a risk of cancer of 1% should trigger investigation or referral.

There have long been concerns, raised by patient groups and supported by evidence from academic studies, that referrals from GPs are sometimes made too late., A recent UK-based study found that, while the median interval between presentation to primary care and referral to secondary care for patients subsequently diagnosed with cancer was 5 days, this varied significantly (interquartile range 0–27) and was significantly faster for some cancers than others.

The causes of late diagnosis of cancer in general practice are still not fully understood. It is often suggested that the ‘gatekeeping’ role of GPs in the UK contributes to the problem. Research using patient vignettes found that GPs in the UK were much less likely to investigate or refer immediately than doctors in countries with higher cancer survival rates.

Efforts to encourage GPs to refer early for suspected cancer can also be met with resistance from commissioners under pressure to limit hospital referrals, and by limited secondary care capacity for diagnostic tests such as endoscopy. Wide variations persist between GP practices and CCGs in the use of the urgent referral route and in so-called conversion rates (the proportion of people referred who are subsequently found to have a cancer). Currently around 37% of all cancers are diagnosed through the urgent (2-week wait) route (as discussed further in section 3.3.4 and shown in Figure 22). A further 25% of patients subsequently diagnosed with cancer are referred non-urgently by their GP. This group experience much longer waits overall to diagnosis and treatment.

3-3-4 Routes to diagnosis: a fuller picture at last?

It wasn’t until 2010, 10 years after the original NHS Cancer Plan, that a more comprehensive analysis was available that could offer a new perspective on how patients with cancer reached the health system, and the subsequent impact on their survival rate. The ‘Routes to Diagnosis’ research project was one of the first major successes of the NCIN (see section 4.4). This linked datasets from screening services, cancer waiting time data, hospital episode statistics and cancer registries, which contain details of cancer type, staging and survival. The research was part funded by the Department of Health, part funded by Cancer Research UK, and was undertaken by the NCIN core team.

In 2010, the first briefing from the research was published, which looked at data from 2006. It contained a surprise: a much higher-than-expected rate of cancers diagnosed as emergencies in hospital, as the project lead at Cancer Research UK recalls:

This other category that we had not even known we were going to find, which was emergency presentations, it was nearly a quarter of all diagnoses coming through an emergency route. Politically, this was obviously seen as totally unacceptable and drove a whole load of initiatives and subsequently has driven metrics around the current cancer plan and cancer programme. One of the key metrics is now reducing the number of emergency presentations.

Sara Hiom, Director of Early Diagnosis and Cancer Intelligence, Cancer Research UK

Not only was the proportion of people presenting as emergencies a shock – nearly a quarter – but the data also revealed that, for some cancers, this was associated with much lower 1-year survival. Figure 22 shows the proportion of cancers diagnosed by each route, and Table 1 how 12-month survival varies by route of presentation for different cancer types.

Figure 22: Proportion of cancers diagnosed by each route for all cancers

All cancers include all Malignant Neoplasms (excluding non-melanoma skin cancer). Percentages shown do not always total 100% due to rounding.

Source: Routes to diagnosis, National Cancer Registration and Analytical Service, Public Health England. Available at: (accessed on 22 October 2018).

Table 1: 12-month net survival rate for breast, colorectal, lung and prostate cancers diagnosed by each route (2011–15 cohort)

All routes

Screen detected

Two-week wait

GP referral

Other outpatient

Inpatient elective

Emergency presentation

































* Female breast cancer only, not including female breast (in-situ). ** Due to a small number of deaths in the screen-detected cohort, 12-month net survival estimates are not available. *** Not applicable – no national screening programme.

Source: Routes to diagnosis, National Cancer Registration and Analytical Service, Public Health England. Available at: (accessed on 22 October 2018).

There has been progress across all cancer types, with more people referred under the 2-week wait and a consistent drop in emergency presentations (Table 2). There are no comparable data of this kind for other health systems. In the case of colorectal cancer, it also shows the potential for screening to improve outcomes (in 2006 the bowel screening programme was yet to start).

Table 2: Percentage of people diagnosed at emergency presentation, 2006 and 2015

Percentage of cancers diagnosed at emergency presentation

Primary cancer site













































The gradual accumulation of evidence about the importance of early diagnosis and its multi-faceted nature has led to something of a policy crossroads around the role of primary care. On the one hand, there are strong arguments for empowering GPs to do more diagnostic work:

An inordinate amount of time is spent chasing missing results so, if I wasn’t sending them away, if I had a point of care test to use with the patient with me, I would be triaging there and then with an immediately available test result, to be able to make a much more informed management decision on whether or not a patient needed a referral or a test that was undertaken in hospital. My analogy is that I have no idea why I can take my dog to the veterinary surgeon and the dog is worked up and managed in half an hour. It has its x-ray and it has its blood test and it’s starting on a treatment, and I cannot do the same for my child.

Dr Fiona Walter, Reader in Primary Care Cancer Research, University of Cambridge

On the other hand, there is a growing belief that the gatekeeping function of general practice might need to be side-stepped completely. In practice, this may need a combination of more diagnostics in primary care (eg FIT for patients with non-alarming bowel symptoms) and increased referral (or routes which bypass general practice) for other conditions, for example low-dose CT scanning for possible lung cancer.

Since 2015, Cancer Research UK and Macmillan Cancer Support have been working with the NHS to explore a range of projects designed to diagnose patients earlier. The Accelerate, Coordinate, Evaluate Programme (ACE) is now on its second wave, exploring the impact of ‘one stop shop’ diagnostic centres. In April 2018, NHS England announced 10 ‘rapid diagnostic and assessment centre’ pilots designed to accelerate cancer diagnosis. These allow GPs and other health care professionals to rapidly refer people with non-specific symptoms which would otherwise not fit existing 2-week wait pathways. For Sean Duffy, (National Cancer Director 2013–15) who worked with the charities to get the ACE programme running, the Department of Health and Social Care-funded evaluation of wave 2 is eagerly awaited to see if it will deliver gains for patients with cancer and other diseases:

This is not just for cancer, because the patients don’t pitch up [at the GP] with a cancer label, they come in with a set of symptoms that you are worried about. At the moment, if it’s a red flag [symptom] you get taken care of, but anything else – I describe this ping-pong of activity that goes on, in and out of different ‘ologys.’ It’s not just about cancer, the same will happen if you’re breathless.

3-3-5 Improving the speed of diagnosis in the hospital sector

Meeting the 2-week waiting time standard required local networks to first map and then redesign how referrals moved between GP practices and hospital outpatient departments. Much of the early work in networks was supported by the Cancer Services Collaborative, to get people in one room and map their services:

I remember having a day with the GPs about getting more rapid access to our outpatients. We identified, I think, it was 24 steps from a GP saying, ‘I’m going to refer,’ to actually being seen. We got it down to three, I think, after a couple of months. My biggest success was getting the pathology turnaround time reduced by a week overnight just by timing when the porters came and picked up the collections from the biopsy suite or bronchoscopy suite.

Mick Peake, Professor of Respiratory Medicine and former Cancer Services Collaborative Lung Cancer Lead

Meeting the other dimensions of the waiting time targets required even more complex work across different hospital departments, including imaging and pathology. Most patients presenting with possible cancer require a combination of clinical assessment, some form of imaging (eg ultrasound, CT or MRI scanning or mammography) and a biopsy to confirm or exclude the diagnosis. For some cancers, biopsy may require an internal inspection of an organ (eg stomach, colon or bladder) via an endoscopy. Once the diagnosis of cancer has been made, further tests may be needed to assess the extent of spread of the disease (staging) and to determine the most appropriate treatment (eg molecular tests to assess likelihood of response to particular drugs). The arrival of new tests, and in particular the expansion in genetic testing of tumours, enable treatments to be targeted at patients who are most likely to benefit. This does, however, mean that, for many cancers, diagnostic pathways are now far more complex, involving a greater range of tests and processes before treatment can begin.

Despite the increasing complexity of the diagnostic process, the targets set nearly 20 years ago in the NHS Cancer Plan are still in force: 1 month between diagnosis and treatment for all cancer patients (the 31-day target; 96% of patients to be treated within this time), and 2 months between urgent GP referral and first treatment (the 62-day target; 85% of eligible patients). Both targets had to be met by 2005. The 62-day target was met and exceeded in 2006 (Figure 23), and maintained until 2013/14 (Figure 24).

Figure 23: Proportion of urgent GP referrals resulting in treatment within 62 days between May 2005 and July 2006, England

Source: Department of Health Cancer Waiting Times Database; Waiting Times for Cancer: Progress, lessons learned and next steps (October 2006). A report to the Secretary of State by Professor Mike Richards.

Despite an injection of capacity, of both equipment and workforce, the 62-day target proved tricky to meet. A progress report published by the Department of Health in 2006 conceded that ‘little overt progress’ had been made between 2000 and 2004. As a result, the efforts of the Cancer Services Collaborative, an Institute for Healthcare Improvement-inspired improvement programme set up in 1999 by the now disbanded Modernisation Agency, were beefed up with much more stringent performance management from the Department of Health from 2004. A national weekly data collection was set up to feed information back to the Department of Health and the Prime Minister’s Delivery Unit.

Figure 24: Urgent GP referrals resulting in treatment and proportion seen within 62 days between Q4 2008/09 and Q2 2018/19, England

Source: Data provided by Cancer Research UK. Source: Cancer Waiting Times – National Time Series – Provider based; NHS England (2018). Available at: (accessed on 23 October 2018).

Laggard trusts were identified for additional ‘support,’ as Janet Williamson former Director of the Cancer Services Collaborative remembers:

We had the ‘naughty 30’. If you remember, the Department of Health team had every single trust and where it was positioned on the grid, and I think that’s probably one of the first times that had ever been done. Well, it was the first time, but there was a huge resource behind that. There were 20 or 30 staff, weren’t there, working on data, and then we had that red list of about 30, but a lot of that came out of that PMDU [Prime Minister’s Delivery Unit], the tracking, and the monitoring.

This heavy-duty performance management worked: in the 2006 report’s own words, progress was ‘spectacular’ from September 2005, with both targets met from July 2006, and maintained thereafter. Trusts whose performance was felt to be lagging were supported by an ‘Intensive Support Team’ (IST), yielding often-dramatic improvements in performance (Figure 25).

Indeed, there was enough confidence about the permanence of the improvement made to this aspect of the pathway that, in 2010, the new Secretary of State for Health, Andrew Lansley, requested that a committee consider whether these standards needed to be retained at all. The committee decided that the targets benefitted patients and should be retained at national level.

Figure 25: Progress resulting from the interventions of the Intensive Support Team on the group of 30 trusts originally selected for support

Note: 62-day performance statistics are only comparable from September 2005. Source: Waiting times for cancer – Progress, lessons learned and next steps; Department of Health (2018).

It is important to note that the 62-day standard does not apply to patients who are referred through a non-urgent route, though the 31-day diagnosis to treatment standard does apply to these patients as well as to those who are referred urgently. The original NHS Cancer Plan set an ‘ultimate goal’ that patients should start treatment within 1 month of urgent referral by 2008, which was ‘in line with the best that patients experience in Europe and the USA’.

The goal for all patients (whether referred urgently or not ) to be diagnosed and treated speedily was revived as a ‘Faster Diagnosis Standard’, recommended in the 2015 strategy, Achieving world class cancer outcomes. This states that all patients should have cancer diagnosed or excluded within 28 days of referral by a GP by 2020. This is now being piloted and data collection methods are being established before full rollout.

Figure 26: International comparison of the number of CT scanners (per million population) in 2015 (or nearest year)

* The OECD average is calculated for the 34 OECD countries for which data were available. ** The UK data used in this comparison are from 2014.

Source: OECD Health Statistics (2017). Available at: (accessed on 22 October 2018).

This new standard is likely to be extremely difficult to meet. Meanwhile, performance against the 62-day standard has continued to deteriorate (Figure 24). As the number of patients referred in to these pathways, and the complexity of the pathways themselves increases, there are concerns that there may not be enough scanning equipment, and workforce to use it. Figures 26 and 27 demonstrate that the UK lags right at the bottom of the OECD in terms of absolute numbers of CT and MRI scanners, ranking 35th out of 37 countries for CT, and 31st out of 36 for MRI.

Unsurprisingly, given the relative paucity of equipment, the UK also lags far behind in terms of absolute numbers of scans done. Although the number of CT and MRI scans being done in the UK has risen steadily over the past 10 years, OECD data suggest 79.3 CT scans/1,000 population occur in the UK compared with the OECD average of 143.1 (ie 80% higher in the OECD than in the UK). For MRI, the UK average of 52.6 MRI scans/1,000 population lags significantly behind the 64.8/1,000 OECD average, 23% higher than the UK (of note, the datasets do not allow us to look specifically at scans for cancer or suspected cancer).

Figure 27: International comparison of the number of MRI units (per million population) in 2015 (or nearest year)

* The OECD average is calculated for the 33 OECD countries for which data were available. ** The UK data used in this comparison are from 2014.

Source: OECD Health Statistics (2017). Available at: (accessed on 22 October 2018).

The lack of endoscopy capacity discussed elsewhere in this report risks becoming a rate-limiting step in the ability to get more people through diagnostic pathways. In 2010/11 crude colonoscopy rates were calculated by the Department of Health and compared to those in Norway, Scotland, Poland, Australia and two provinces in Canada (Table 3).

Table 3: Colonoscopy rates per 1,000 population 2010/11











Alberta (Canada)


Nova Scotia (Canada)


Source: Unpublished analysis by the Department of Health.

Figure 28 shows a gradual increase in all-cause endoscopy numbers since 2007, and anecdotally we were repeatedly told that this slow rise has not kept pace with increased demand.

Figure 28: Number of colonoscopies, flexi sigmoidoscopies, cystoscopies and gastroscopies completed between 2007 and 2017

Source: Monthly Diagnostics Data; NHS England. Available at: (accessed on 22 October 2018).

Slippage against the 62-day target (not consistently met in England since 2013/14) is likely to be due to a combination of factors, including the steadily rising numbers of urgent referrals, but also the cumulative effect of system change and funding restrictions since 2013.

3-4 Better cancer treatment

There are three main approaches to treatment of cancer: surgery, radiotherapy and systemic therapies (which include chemotherapy, hormonal therapies and newer biological/immunological treatments). All three approaches have improved over the past 20 years. They are often used in combination, for example surgery plus radiotherapy plus chemotherapy/hormonal therapy for many patients with breast cancer. They can also all be used either with curative or palliative intent. For example, surgery can be done to relieve bowel obstruction and thereby improve symptoms, even when it is clear that this will not be curative. Similarly, radiotherapy for a bone metastasis can be very effective at relieving pain, but will not provide a cure.

Surgery cures more patients with cancer than the other methods, and is often the least expensive. Chemotherapy is often thought (by non-experts) to have the greatest impact on survival, but overall this is not the case. It does, however, vary by cancer type. For leukaemias, lymphomas and testicular cancers, chemotherapy is of central importance.

The role of adjuvant therapy (usually given after surgery to reduce the risk of relapse) has grown markedly in the past 20 years for breast and colorectal cancers – and some others. This has been in response to evidence from clinical trials of improved survival.

One of the key problems 20 years ago was that patients were not being assessed by teams with expertise in all of these methods and were therefore missing out on treatments. This was the prime reason for establishing MDTs. For surgery, there was an additional problem that some patients requiring complex surgery were being treated by surgeons who only rarely performed that surgery.

Although Calman–Hine recognised these problems (and hence recommended centres and units), Kenneth Calman also recognised that much more detailed guidance was needed on individual cancers. This led to the Improving Outcomes Guidance programme, which was supported by rigorous evidence reviews.

The NHS Cancer Plan did not go into detail about the different types of cancer treatments and how they were to be improved. The focus instead was on implementation of the approach outlined in Calman–Hine, providing a timetable for producing service guidance for each cancer type, rolling out peer review across the country to underpin scrutiny of the guidance, and, via the cancer networks, ensuring that all patients were seen by multidisciplinary cancer teams. The recommendations about the most up-to-date (and cost-effective) cancer drugs now became the responsibility of NICE.

3-4-1 Multidisciplinary teams: the cornerstone of high-quality care

Good decision making on cancer treatments almost always requires input from several different specialists, together with good communication with, and involvement of, patients. Although the combination of specialists required varies between cancer types, most commonly this involves surgeons, clinical oncologists, medical oncologists, pathologists, radiologists and nurse specialists working together as a team, with administrative support.

Before the Calman–Hine report (1995) this form of MDT working for cancer was the exception rather than the rule. MDTs had existed in some specialist centres for decades, but the large majority of patients were not benefitting from this approach. MDT working became formalised for breast cancer before other cancers, partly as a result of the introduction of breast screening in the late 1980s. However, despite this, the British Breast Group (a multidisciplinary group of experts in different aspects of breast cancer) felt the need to publish a report in 1994 making the case for MDT working becoming the norm.

Although advocated by the Calman–Hine report, in 2001 the Commission for Health Improvement reported that progress on MDT working in the years between 1995 and 2000 had been patchy.

Improving Outcomes Guidance reports were subsequently used as a vehicle for encouraging MDT working, with each of those published between 1996 and 2005 re-emphasising the need for MDTs and specifying the range of specialists required in them. These requirements were then reflected in the standards used for the national cancer peer review programme (described more fully in section 4.3). This combination of evidence-based guidance, peer review and MDTs is seen by many as having made a crucial difference to the quality of cancer care.

There’s no question but that the widespread adoption of an MDT-based system of management has led to improvements of practice across the board. I mean, sad to say that the most important change is probably stopping maverick clinicians doing what they had decided, on very little evidence, was the best thing to do for this patient because that’s what they always did for this patient.

Professor Sir Alex Markham, Professor of Medicine, University of Leeds, and former Chief Executive of Cancer Research UK

It enforced non-tribal working. It looked at whole pathways of care. It’s made two groups of people in particular perform in front of an audience, where the quality was woeful. That is, radiologists and pathologists. It frankly made people raise their game, because otherwise they were embarrassed on a weekly basis. So, rapid change of quality of both reporting and relevance to clinical decision making.

Adrian Crellin, Consultant Clinical Oncologist, Leeds

There are now around 2,000 cancer MDTs in England, with the vast majority of patients being managed by such teams. The workload associated with reviewing every cancer case at a formal MDT meeting is now being reviewed. In some cases, MDTs can be over-burdensome for staff, and approaches that streamline processes while retaining the benefits of multidisciplinary input are now needed.

3-4-2 Surgery

Calman–Hine recommended that complex cancer surgery should be undertaken in specialist centres and that reconfiguration would be needed. The Improving Outcomes Guidance provided more detailed evidence reviews to support this and recommended numbers of procedures that should be done in centres. This had big implications for the workforce and local hospitals, as the architect of the Improving Outcomes Guidance remembers:

Roles would become more specific. The generalists would stop fiddling around with stuff and the specialists would do more of that work.

Bob Haward, Professor of Cancer Studies, University of Leeds

The approach for each cancer was set out in the respective Improving Outcomes Guidance. Implementation was the responsibility of cancer networks and progress was monitored through peer review, with oversight by the National Cancer Action Team.

Some types of surgery needed more change than others. While breast and colorectal remained largely based in district general hospitals, lung and brain surgery were already specialised. For breast cancer, surgery in district general hospitals reflected the large volume of patients requiring surgery, while for colorectal cancer the need to have surgeons available to deal with non-cancer-related emergency abdominal surgery ensured that colorectal surgeons remained in district general hospitals. Surgery for prostate, bladder and oesophagus/stomach needed the biggest overhaul, along with some head and neck, and gynaecological procedures.

As an example, in 1997/98, 17% of surgery for prostate or bladder cancer was done in trusts doing fewer than nine procedures a year. In 2006/07 that had dropped to 2%, whereas 77% of surgery was done in trusts undertaking more than 40 cases a year (Table 4).

Table 4: Change in prostatectomy and cystectomy procedure volume by location over time

% of procedures taking place in trusts undertaking:



1–9 procedures per annum



10–39 procedures per annum



40+ per annum



Source: Cancer Reform Strategy: first annual report (2008).

Driving these shifts was the evidence assembled in the Improving Outcomes Guidance. The presence of a credible evidence base, according to Professor Bob Haward, was crucial in reconfiguration:

We were able to propose quite radical reconfigurations which the world broadly accepted. I mean, there were fights here and there about particular places and how it would be interpreted in this part of the country or that. I remember being called before the national committee of the clinicians involved in upper GI. Allegedly we were going to have our balls removed and hung out to dry. As it was, we came into a room full of people who were lying on their backs with their paws in the air and, in a way, it demonstrated that, really, it was hard to fight it and they, in the end, didn’t.

Bob Haward, Professor of Cancer Studies, University of Leeds

For some cancers, particularly lung cancer, the focus has been on increasing the number of patients receiving surgery. This work has been led by chest physician Mick Peake, who remembers a ‘completely and totally nihilistic attitude’ towards lung cancer among fellow clinicians:

A very prevalent view was that this [lung cancer] was all smoking related and self-inflicted, and they all died anyway, and it’s nothing to do with the chest physicians. So, I just felt it needed fixing really. It was an obvious clinical need that hadn’t been addressed.

Armed with evidence from peer review and subsequently from the National Lung Cancer Audit that patients with operable disease were often not being referred for surgery, Mick Peake demonstrated that it was possible to triple the surgery rate for lung cancer (in one cancer network) within a year, with the help of the Cancer Services Collaborative improvement team. Spreading these improvements across all cancer networks was catalysed by the release of the first lung cancer audit in 2005. Between 1985 and 2005, there were, on average, about 3,000 lung cancer operations a year.

In 2016, it was 7,250, and 2017’s going to be above that. So, the number of operations has more than doubled. They’re doing complex procedures. The average age of those being operated on has increased, and the mortality rate has dropped at the same time. So, they’re dealing with an older, less fit, more complex, more advanced stage surgery, and the mortality rates are now below anything international.

Mick Peake, Professor of Respiratory Medicine and former Cancer Services Collaborative Lung Cancer Lead

Surgical technique has also evolved. Surgery is now undertaken in situations that would not have been possible 20 years ago, for example liver resections (to remove metastases from colorectal cancer in the liver). There have also been major developments in minimally invasive forms of surgery, including sentinel node biopsy in breast cancer, and laparoscopic surgery for colorectal cancer (the adoption driven by a national training programme, underpinned by the bowel cancer audit). In breast cancer, the practice of surgery has changed, moving away from mastectomies towards breast conservation and reconstruction.

Another major surgical change has been breast reconstructive surgery, which has gradually developed over the last 20 years to the point that, at the Royal Marsden Hospital, 50% of our patients choose breast reconstruction at the time of mastectomy. We’re at the point where we’re using the biology of the cancer more than the disease extent and stage, to guide the initial treatment. Treating the cancer with drugs such as chemotherapy, Herceptin and Letrozole as the first treatment has the potential to allow us to de-escalate from radical surgery such as mastectomy and axillary clearance to more conservative surgery preserving the majority of the breast and axilla. There are increasing numbers of women who will not benefit from radical surgery if the cancer shrinks or disappears on the initial drug treatment. So, in the future, the next decade or so, we will be looking at the role of surgery and its sequencing with the other cancer treatments so we get the best survival outcomes for minimal surgery and complications.

Fiona MacNeill, Consultant Breast Surgeon, Royal Marsden NHS Foundation Trust

3-4-3 Radiotherapy

Radiotherapy is a vital component of cancer treatment, and has been used since 1899. It can both help cure and alleviate symptoms. By the end of the 1990s, it was clear that there was an inadequate provision of radiotherapy machines and staff to operate them. This was a legacy of underinvestment in a sector that had once looked as though it might be made redundant as an effective treatment of cancer. As a leading cancer specialist remembers:

You know, when I started [my career] in 1985 a senior colleague called me aside and said, ‘Are you really sure you want to do this? You do realise that everything’s going to be cured by chemotherapy and that radiotherapy is a complete waste of time.’ In the 1970s and 80s, various figures had very clearly signalled the death of radiotherapy and it, therefore, became a Cinderella thing, underinvested, and the quality actually went. We just lost all momentum to keep pace with what was going on in the rest of the world, which was a massive acceleration of complexity and technology.

Adrian Crellin, Consultant Clinical Oncologist, Leeds

Radiotherapy has consistently had a much lower public profile compared with chemotherapy and access to cancer drugs, which has had a powerful pharmaceutical industry behind it, and attracted much more media interest. Nevertheless, there have been important improvements in radiotherapy techniques. Examples of selected innovations are given in Box 1, all of which provide more targeted treatments, with fewer side effects for patients.

Box 1: Innovations in radiotherapy since 2000

Innovations in radiotherapy since 2000:

  • Intensity-modulated radiotherapy
  • Image-guided radiotherapy
  • 4D adaptive radiotherapy
  • Stereotactic ablative radiotherapy (also known as stereotactic body radiotherapy)
  • Proton beam therapy

While the number of episodes of radiotherapy has broadly increased over time, the trend has levelled off in the past 2 years (Figure 29). The proportion of patients receiving intensity-modulated radiotherapy is currently increasing year on year, from 32% in 2014/15 to 46% in 2017/18.

There have been several pulses of central capital investment in equipment, the largest in 2000, and from 2015 in terms of volume of new machines (Table 5). Radiotherapy is administered by a machine known as a linear accelerator, or linac. Linacs last around 10 years and then need to be replaced. Expanding radiotherapy therefore needs both a rolling programme of renewal and upgrading, alongside the purchase of new machines.

Figure 29: Radiotherapy activity in England, 2009/10 to 2017/18

Data provided by National Cancer Registration and Analytical Service, Public Health England. Source: Radiotherapy Dataset (RTDS), NCRAS, Public Health England.

Table 5: Central capital investment in equipment


Funding/numbers of machines

NHS Plan/Cancer Plan


106 linacs (cost unknown)

Innovation Fund



Proton beam therapy


£230m for two centres

2015 Cancer Strategy


£130m for new and renewed linacs

In 2000, the NHS Plan promised 50 new linacs, and the NHS Cancer Plan a further 56, funded through the National Lottery. National oversight of investment and implementation rested with the National Cancer Action Team, assisted by a National Radiotherapy Advisory Group and an implementation group.

Despite the investment in 2000, a report commissioned by the Department of Health and published in 2007 concluded that the pace of the growth in demand for radiotherapy had been underestimated and that more capacity was urgently needed. The report estimated a 63% gap between existing and optimal radiotherapy levels for patients, and an ‘unacceptable’ variation between regions. It called for a 91% increase in activity to meet demand, by 2016. The report also highlighted that action to train more therapeutic radiographers since 2000 had been hampered by high attrition rates (ie students not completing the training), of 35%.

The 2007 Cancer Reform Strategy did not allocate additional investment for radiotherapy, but it was hoped that the inclusion of radiotherapy in the 31-day treatment target would focus the attention of local hospital trusts and commissioners on the need to invest locally in adequate services.

In 2012, the government committed £230m to build two proton beam therapy centres, in Manchester and London, to treat a minority of patients who would otherwise have had to travel abroad. (The first machines are due to begin work in 2018.) The most recent tranche of investment (£130m) was made following the 2015 strategy (Achieving world class cancer outcomes), to renew and replace linacs across England.

Two themes emerged from interviews. First, it has been difficult to incentivise trusts to invest consistently in renewing equipment, and a second (related) theme is the challenge of spreading innovative techniques. While national action has been important to kick start the use of new techniques, such as intensity-modulated radiotherapy and proton beam therapy, some interviewees questioned why national bodies are still expected to identify gaps and provide funding for routine equipment renewal in trusts, despite the creation of a national tariff for radiotherapy activity, which includes an element to cover depreciation.

Radiotherapy services have suffered through lack of investment in the latest technology and due to trusts not seeking to ensure that funding from the radiotherapy tariff is kept aside for essential equipment replacement. Whilst government handouts of funding are welcome in rectifying this problem in the short term, it doesn’t encourage trusts to plan appropriately for the next round of equipment replacement. Either equipment should be managed, ie funded at the national level, or trusts incentivised to ensure they plan for replacement on a regular and rolling basis.

Charlotte Beardmore, Director of Professional Policy, Society of Radiographers

Expanding access also depends on recruiting additional staff. But the adoption of new technologies seems to have compounded an already challenging workforce planning task, complicating some tasks (and requiring better treatment planning), while simplifying others:

The technology is getting ever better, and you can bend and shape a beam in whatever direction you like these days. It’s much more technically demanding so some radiotherapy now has been pushed off to radiographers, the simple breast post-lumpectomy radiotherapy, which is pretty straightforward. But planning in 3D and with CT and MRI scanners obviously takes time.

Adrian Crellin, Consultant Clinical Oncologist, Leeds

In early 2018, an All-Party Parliamentary Group (APPG) for Radiotherapy was formed and published a ‘manifesto.’ This highlighted the relatively low spend on radiotherapy (5% of the cancer treatment budget) and called for investment in new radiotherapy equipment, to bring the UK in line with some other comparable countries (Figure 30).

Figure 30: International comparison of radiotherapy equipment (per million population) in 2017 (or nearest year)

Source: OECD (2018), Radiotherapy equipment (indicator). Available at: (accessed on 15 October 2018).

3-4-4 Chemotherapy and other systemic treatments

Chemotherapy, used on its own and in combination with other forms of treatment, is the third vital component of cancer treatment. It is by far the most expensive, and access to chemotherapy has been transformed since 2000. There has been very significant improvement in the use of what might now be called ‘old’ drugs and the development of novel, targeted treatments.

A far higher proportion of patients with cancer are now treated with drugs than 25 years ago, although this is difficult to quantify as good data on chemotherapies were not collected until recently. The increase is partly because chemotherapy is now being used for some cancers for which this was previously rare (eg cervix, prostate, non small-cell lung cancer, oesophagus and stomach). In addition, expansion in the drugs available means that individual patients who might only have received one or two different regimes in the past might now receive five or more – gaining a response each time, but with the disease then progressing again.

The importance of improving access to cancer drugs pre-dated the NHS Cancer Plan. After the Downing Street cancer summit in 1999, 13 drugs were earmarked for urgent appraisal by the new NICE. A pivotal moment was the decision, in 2001, to make it mandatory for PCTs to fund positive NICE technology appraisals of new cancer drugs. Once a drug was approved by NICE, local commissioners had 3 months to ensure it was funded, and money was allocated to commissioners at a national level to enable this.

This led to an expansion of drugs available to cancer patients, at the same time as shifting, for a while, the political heat away from ministers about which drugs should be available.

The creation of NICE did not depoliticise chemotherapy for long. Pressure on NICE grew from the mid-2000s for patients to get access to drugs that had not yet been appraised by NICE, or had been turned down (see Box 2). In 2009, NICE relaxed its thresholds for certain patients at the end of life, but improving access to cancer drugs had become a campaign issue. In 2010, the new coalition government fulfilled a manifesto promise by announcing the creation of a Cancer Drugs Fund to allow patients access to drugs that were not routinely available on the NHS. In 2015, the NAO reviewed the fund: £968m had been spent on drugs for 74,380 patients between 2010 and 2015. It concluded that it is likely that the Fund has ‘contributed significantly to the improvement in the UK’s relative position in providing access to newer cancer drugs.’ But crucially, the NAO noted, despite this spending, there was no data to assess what impact these drugs had had on patient outcomes, such as survival, making it hard to assess what the benefits might have been if the money had been invested in other forms of treatment.

Box 2: The Herceptin story

Though many new cancer drugs have appeared in the past 20 years, few have had the impact of Trastuzumab – a targeted cancer drug, best known by its brand name Herceptin. Usually used to treat breast cancer expressing a specific protein, HER-2 (present in about 20% of breast cancers), Herceptin was first approved for medical use in the US in 1998. Thereafter followed a bumpy journey in to the NHS: an early test of Labour’s commitment to end postcode lotteries and of the fledgling machinery of NICE built to do just that.

Herceptin had been approved for use in the NHS in 2002, but only for metastatic breast cancer – a life-prolonging but not a life-saving intervention. By early 2005 though, the cancer research community was clear that the uses for Herceptin extended beyond patients with an already terminal diagnosis. In October 2005 the New England Journal of Medicine published the results of three trials designed to test the role of Herceptin as an addition to surgical treatment for primary breast cancer. Described as ‘simply stunning’ in an accompanying editorial, the case for an expanded rollout of Herceptin was becoming more evidence based.

In England the Secretary of State, Patricia Hewitt, moved quickly to make it clear that she wanted wider use of Herceptin to be approved for the NHS as a whole. Inconvenient then that the manufacturer, Roche, hadn’t even applied for a licence yet, a necessary step to enable an application to be submitted to NICE for approval. Hewitt made clear that NICE should ‘fast-track’ the drug as soon as an application was made, and encouraged PCTs to consider funding the unlicensed drug in the interim ‘because I think it is the right thing to do’. Such an intervention was, as the King’s Fund pointed out, ‘a substantial deviation from the procedures set up to recommend therapies for use in the NHS’.

This deviation though, had occurred in the face of unprecedented pressure from charities, patient groups and the media. A campaign group, Women Fighting for Herceptin, had been formed in 2005 with the aim of getting Herceptin free on the NHS for all who might benefit from it. A petition with more than 34,000 signatures was taken to Downing Street and the campaign group met with Minister for Health Rosie Winterton and received significant media coverage in print, on BBC Breakfast and on Woman’s Hour on Radio 4. As one journalist, then health editor at The Independent commented:

‘a story about breast cancer would sail into the paper…. you have a young mother dying, there’s almost nothing – with young children there’s almost nothing – that trumps that in terms of tugging the heartstrings. It’s a dreadful position’.

Cancer charities including Cancer Research UK, the Breast Cancer Campaign and Breakthrough Breast Cancer lobbied loudly, both in the corridors of Westminster and through copious column inches. When Ann Marie Rogers took Swindon PCT to a judicial review to challenge its refusal to give her Herceptin on the NHS in 2006, she lost the case, but won the media battle. Less than 6 months later, in August 2006, NICE licensed Herceptin for use as an adjuvant treatment in early stage breast cancer.

Data are now available from the Systemic Anti-Cancer Therapy Dataset on the numbers of patients treated in England and what was given for different cancers. In the 12-month period from March 2017 to February 2018:

  • 212,344 different patients are recorded as having chemotherapy.
  • 330,474 regimens were given (eg one patient can have two regimens, such as first and second lines of treatment and some therapies have two separate phases with different regimens).
  • There were 1,160,551 cycles with 2,968,925 separate drugs within these cycles.
  • Of these 212,344 different patients, chemotherapy was given for specialties in the following proportions:
    • breast 19%
    • urology 12%
    • lower GI 11%
    • lung 9%
    • lymphoma 8%
    • upper GI 7%
    • myeloma 6%
    • gynaecology 6%
    • leukaemia 5%
    • head and neck 2%
    • skin 2%
    • brain 1.5%
    • paediatrics (0–16) 1.3%
    • teenagers and young adults 0.8%
    • sarcoma 0.7%
    • miscellaneous 8%.

There has been a large increase in spending on chemotherapy. In 2017 around £2.5bn was spent on cancer drugs and their delivery, approximately 2% of the NHS budget. Chemotherapy is currently funded and managed as part of specialised commissioning in NHS England, and represents its single biggest cost pressure.

As regards the current £2.5bn expenditure by NHS England on chemotherapy: if you’d said this figure to me 5 years ago I’d have laughed. If you’d said that to me 10 years ago I’d have said ‘which country are you talking about?’. We have this because of the mandatory funding direction from NICE. Within NHS England Specialised Commissioning the expenditure on chemotherapy swipes four-and-a-half times the size of the nearest other single expenditure which is radiotherapy, which is £450m to £500m.

Professor Peter Clark, Chair, NHS England Chemotherapy Clinical Reference Group

Chemotherapy has, in some cases, revolutionised the lives of patients in the opinion of those we interviewed (see Box 3). The creation of a good-quality dataset, that can answer questions about the relationship between all the cancer drugs in use (rather than just the minority accessed through the Cancer Drugs Fund), alongside their costs and outcomes, has been slow:

I’m frustrated that we haven’t yet got the routine analysis of outcomes, you know, proper, meaningful outcomes such as treatment duration and survival. Well, for CDF [Cancer Drugs Fund] drugs where we’re investing heavily in SACT [Systemic Anti-Cancer Therapy] and PHE [Public Health England], we know we’re going to get those data and are beginning to get them now. For the routine access and outcome data, I still think it’s going to take another year or two.

Professor Peter Clark, Chair, NHS England Chemotherapy Clinical Reference Group

Box 3: Some of the ‘game changers’ in systemic treatments according to interviewees

Tamoxifen – Arguably the first game-changing systemic therapy, tamoxifen was discovered in 1962. It wasn’t until the early 1990s that a meta-analysis from the Early Breast Cancer Trialists’ Collaborative group conclusively showed that tamoxifen saved lives in early breast cancer. Since then it has become the mainstay for treatment of oestrogen-receptor positive breast cancer.

Trastuzumab – Approved by NICE in August 2006, trastuzumab (Herceptin) is used to reduce the risk of recurrence in breast cancers expressing the HER-2 protein. Usually given for 12 months, it can also be used in advanced HER-2 positive breast cancer to slow disease and extend survival.

Imatinib – Approved by NICE in October 2003, imatinib (Glivec) rapidly became the most common treatment for chronic myeloid leukaemia. A daily tablet, taken (often) indefinitely, Glivec has turned this into a disease that, in most cases, can be controlled for many years.

Rituximab – Approved by NICE in September 2003, rituximab has significantly improved survival for people with non-Hodgkin lymphoma as well as being used as a treatment for a variety of autoimmune diseases such as rheumatoid arthritis.

Pembrolizumab – a type of cancer immunotherapy which stimulates the immune system to fight cancer cells. It has recently been approved by NICE for metastatic melanoma, non-small-cell lung cancer, relapsed or refractory Hodgkin lymphoma and some advanced bladder cancers.

3-4-5 Patient experience and living with and beyond cancer

If you’re told you’ve got a cancer, there are three crucial questions, ‘Am I going to survive?’ ‘Am I going to be treated well?’ and ‘What am I going to be like afterwards?’ When we assess outcomes in cancer, in general, the easiest and most objective parameter, is ‘Am I going to survive?’

Adam Glaser, Professor of Paediatric Oncology, University of Leeds

This section deals with how well the cancer strategies have addressed the second and third of the questions above: patient experience, and living with and beyond cancer. An estimated 2.5 million people were living with cancer in the UK in 2015. This is predicted to rise to 4 million people by 2030. This reflects both the increasing numbers of new cancer cases, and improvements in survival in recent years.

Patient experience

The idea that patients’ experience of care is as important as their clinical outcomes seems uncontroversial now. But many clinicians interviewed for this report remember a time when providing patients with even the most basic information about their cancer did not always happen, as oncologist Adrian Crellin remembers from the early 1990s:

I was working in Pontefract where one woman, a lovely little old lady, really smashing, no faffing around, [said] ‘You look like a nice young man? I hope you don’t mind me asking, but is it cancer? I thought it probably was when I woke up from the operation and the breast was gone, but I didn’t want to bother the surgeon because he seemed to be in a rush and I didn’t want to trouble him. You look like a nice young man, I hope you don’t mind me asking.’ Those were the words that were used. I’ve never forgotten it. It was one of the most shocking consultations I think I’ve ever been in.

From the 1980s onwards, evidence emerged about the value of good information and psychological support for cancer patients, and the importance of good communication skills on the part of their clinicians. Patient charities, particularly CancerBACUP, had lobbied hard for information and support to be a normal part of treatment. In 2000, the NHS Cancer Plan promised a national patient survey for cancer, originally promised by Frank Dobson (then Secretary of State for Health) in 1998, to run alongside a similar survey for heart disease.

The first survey, of 65,000 patients from 172 trusts, was published in 2002, followed by another in 2004. Regular annual surveys then followed from 2010, known as the National Cancer Patient Experience Survey. Data are published nationally, but are also presented back to hospital trusts, with performance against individual questions marked in red, amber and green.

It was a powerful lever for trusts, they couldn’t ignore that, well they would try to sometimes, but it was hard to ignore the data around patients’ reports of poor experience. It was forensic and the whole traffic-lighting was very powerful in terms of being able to work at a granular level, at a trust level, even down to a team level, to drive change. It was hugely powerful.

Amanda Ramirez, Professor of Liaison Psychiatry, King’s College, London

In addition to the survey, the NHS Cancer Plan set up communication training for clinicians (evidence for which had been provided by a trial, led by Professor Lesley Fallowfield) and a review of information available to patients. There was ample ground to cover: in 2002, a review of cancer services by the Commission for Health Improvement and the Audit Commission found that only a quarter of consultants reported having had any communication skills training, and the written information available to patients was highly variable, with some only offering a few leaflets or none at all.

One of the components of a positive patient experience has been the presence of CNS, who provide cancer patients with information, support and continuity, in addition to clinical care. Their numbers have risen over the years, but it remains the case that not all cancer patients have access to a CNS. Research by Breast Cancer Care found that not all patients with metastatic breast cancer have access to a CNS, despite these being patients who may need the most support. The charity Macmillan Cancer Support has played a pivotal role in spreading CNS posts across the NHS. Macmillan has pump-primed CNS positions for 3 years, in return for continued NHS funding thereafter.

By 2011, the National Cancer Patient Experience Survey was able to shed light on whether important aspects of care were in place. For example, it demonstrated whether surveyed patients were actually receiving help from a CNS, and analysis showed that patients who received support from a CNS had a better experience. The survey also asks questions about whether patients were given their diagnosis in a sensitive way, provided with written information about their treatment, had side effects explained and were offered help managing symptoms.

Living with and beyond cancer

By the time of the second cancer strategy, in 2007, the quality of life for people who had survived cancer and completed their treatment came into focus, as research emerged about the poor quality of life reported by patients who had had bowel cancer and prostate cancer in particular. In 2008, the first estimate was made of the number of people living with cancer in the UK: 2 million (expected to grow by 3% a year), but the evidence base was limited about what interventions would best address people’s needs. A National Cancer Survivorship Initiative (NCSI) was set up, in collaboration with Macmillan Cancer Support (which successfully campaigned for the abolition of prescription charges for cancer patients in 2008) and other NHS partners. The NCSI set out to build a better evidence base and develop new models of care to improve the health and wellbeing of cancer survivors.

The initiative produced a ‘vision’ for better survival services in early 2010. This called for a significant shift in the culture of care, with a greater focus on recovery and wellbeing after treatment. This had to be tailored to each patient, with an emphasis on enabling patients to self-manage, with more targeted follow-up, rather than automatic outpatient appointments.

The overall impression from interviewees was that supportive care had, on the whole, improved over the past 20 years:

Supportive care around patients has changed enormously over the years. We still have pockets of good and less good now, but there’s a lot more about the psychological support of the patients, a lot more around thinking of managing symptoms, managing side effects, about patients and families.

Amanda Shewbridge, Macmillan Nurse Programme Lead for Living With and Beyond Cancer for the South East London Accountable Care Network

But it is difficult to provide any evidence for this. The 2010 vision called for the development of systematic measurement, in the form of a patient-reported outcome measure (PROM) for cancer. A PROM was developed and tested in 2011 for breast, colorectal, prostate cancer and non-Hodgkin’s lymphoma. The national rollout of the PROM was a casualty of the 2012 reorganisation, when its earmarked £2m funding was reallocated in NHS England after 2013.

Work on a quality-of-life measure has now restarted and is being piloted in five areas, with rollout scheduled for 2019. The metric may help assess the impact of the key interventions recommended by the NCSI. These include a recovery package (consisting of a holistic needs assessment, treatment summaries to be shared with GPs, a review with either a GP or practice nurse, and access to health and wellbeing support). The current 2015 strategy wants the recovery package to be available to all by 2020.

Despite the obvious importance to patients, many felt it was difficult to make the case for additional funding for these services in a constrained environment, and that it had slipped down the priority list for policymakers.

The NHS clearly is resource-stretched, or resource-poor. I think the supportive care was seen as a bit of a soft option. There are key bits. You’ve got to answer question one, which is: will I survive? They’ve got to have their operation in time, their diagnosis in a timely manner, get their operation and treatment started in a timely way, and deliver safe radiation and chemotherapy. Undoubtedly, that’s got to be right, because, if you don’t get point one right, you can forget point three [what am I going to be like afterwards?].

Adam Glaser, Professor of Paediatric Oncology, University of Leeds

For charities such as Macmillan Cancer Support, the logic of more people surviving cancer will mean that this dimension can no longer be ignored:

I think we, in cancer, still live very strongly in cell biology, and in clinical mode. If one in two people in our lifetime are going to have cancer, and a lot of people do survive, there is a need for more investment in that area. Whilst survival is ‘king’, how people survive and get back to best quality of life is critical, both for the individual as well as the overall economy.

Fran Woodard, Executive Director of Policy and Impact, Macmillan Cancer Support

Previous Next