Cancer survival rates

We explore how cancer survival varies by cancer type and how it compares across countries.

Qualitywatch

Indicator

Last updated: 27/06/2024

Background

Cancer survival is one of the key measures of the effectiveness of cancer services. Survival rates capture not only how well the healthcare system can detect the disease but also whether people have rapid access to effective treatment. In this indicator, we explore cancer survival for common cancer types and compare the UK's five-year survival rates for breast, cervical and colon cancer with other countries. We also look at trends in the stages at which cancer is diagnosed and unpack inequalities in early cancer diagnosis when categorised by levels of deprivation.

Please visit Cancer Research UK to learn more about cancer survival, and Macmillan Cancer Support for information and support on managing prognosis.


Five-year cancer survival

Here, we look at the five-year net survival estimates for adults (aged 15–99 years) in England diagnosed with one of the 25 most common cancers between 2016 and 2020 and followed up to 2021. There is currently a huge variation in survival between cancer types, resulting from a range of patient-level, treatment and biological factors.

The cancers with the lowest five-year survival estimates were pancreatic cancer (8.3%), brain cancer (12.9%) and liver cancer (13.4%). The highest five-year survival estimates were seen in patients with testicular cancer (93.5%), melanoma of skin (92.6%) and prostate cancer (88.5%). As reported by NHS England, the five-year survival of all patients diagnosed in 2016 was 55.7%, a 7.8 percentage point improvement since 2005.

Some of these patterns may be explained by differences in the stage at which the cancer is diagnosed. In 2021, 43% of all cases of pancreatic cancer (across all routes of diagnosis) were diagnosed at stage 4, when compared with 18% of all cases of prostate cancer. Diagnosis at an earlier stage of a cancer’s development is associated with improved outcomes and increased chances of survival. Early diagnosis can be improved by reducing cancer waiting times, such as the time people wait to see a specialist after an urgent referral from a GP, or the time people wait for diagnostic tests. It can also be improved through public health interventions, such as screening programmes and education campaigns.


Early cancer diagnosis over time

The stage at which a cancer is diagnosed describes the size of a tumour and how far it has spread from where it originated. The numerical staging system classifies cancers into four stages, with stage 1 as the earliest stage. At both stage 1 and 2, the cancer has not spread to anywhere else in the body, although stage 2 indicates that the cancer has grown. If cancer is diagnosed during stage 3, it’s likely that it has spread, at which point the chances of survival decrease as disease prognosis worsens.

This chart shows data on eight common cancers: prostate, breast, lung, colorectal, urological, gynaecological, melanoma and oesophago-gastric. In 2018 and 2019, the number of cancers diagnosed at each stage remained roughly constant, with an average of 42% of cancers diagnosed at stage 1 or 2. Between March and May 2020, following the onset of the Covid-19 pandemic, the total number of cancers diagnosed fell by 36% from 18,077 to 11,566. Over the same period, the proportion diagnosed at stage 1 or 2 decreased from 44% to 36% and the proportion diagnosed at stage 4 increased from 15% to 21%.

Since then, the number of cancers diagnosed has increased to exceed the numbers recorded before the pandemic. In January 2023, 19,169 cancers were diagnosed and 38% of cancers were recorded as being diagnosed at stage 1 or 2. While the proportion of cancers diagnosed at each specific stage has remained relatively stable over time, the proportion of cancers recorded as being diagnosed at an unknown stage has been increasing over time; in January 2023, 34% of diagnosed cancers did not have a stage recorded (see ‘About this data’ for more information).

The NHS Long Term Plan set a target to increase the percentage of cancers diagnosed at stage 1 or 2 to 75% by 2028. Commitment to this target has been reiterated in the NHS priorities and operational planning guidance published in January 2023. Targets for faster cancer diagnosis have also been set which state that by March 2024 75% of patientsshould be told whether or not they have cancer within 28 days of an urgent referral from their GP or a cancer screening programme. The faster diagnosis standard was met in March 2024 with 77.3% of patients given their cancer diagnosis or having cancer ruled out after 28 days of referral (data not shown).


Early cancer diagnosis by deprivation

Early-stage diagnosis comprises cancers diagnosed at stage 1 or stage 2 of the disease progression. Individuals living in the least deprived areas have had a consistently higher rate of early cancer diagnosis than those living in the most deprived areas. In 2018 and 2019, the number of cancers diagnosed at an early stage for both groups remained roughly constant, with slightly more variation occurring for individuals living in the most deprived areas than for those living in the least deprived areas.

Between March and May 2020, following the onset of the Covid-19 pandemic, the percentage of cancers diagnosed early fell from 58% to 52% for individuals from the least deprived areas. The decline for the same period was sharper for individuals from the most deprived areas as early diagnoses fell from 53% to 41%. Since then, the percentage of early cancer diagnoses has increased to levels slightly higher than they were before the pandemic for both groups. In January 2023, 62% of cancers were diagnosed early for individuals in the least deprived areas and 53% were diagnosed early for individuals in the most deprived areas.

Early cancer diagnosis rates for both groups are less than the 75% target set for 2028 by the NHS Long Term Plan. However, it should be noted that individuals living in the least deprived areas are and have always been closer to the target than individuals living in the most deprived areas. Guidance published in February 2022 particularly emphasises the importance of early cancer diagnosis for individuals living in disadvantaged areas


International comparisons of five-year cancer survival

Breast cancer is the most prevalent cancer in women worldwide. Survival is related to advances in treatment, detection through screening programmes and a greater awareness of the disease. Five-year survival for breast cancer has been improving in the UK over time, reaching 86% in 2010-2014.

However, despite the relatively high breast cancer screening coverage when compared with several other countries represented in the Organisation for Economic Co-operation and Development (OECD) data, the UK continues to lag in terms of survival. There is recognition that in terms of cancer mortality, the UK performs worse than other comparable countries, and this may result from differences in access to care and delays in treatment or could be due to population-level factors. The United States reached the highest five-year breast cancer survival rate of 90.2% in 2010-2014, followed closely by Australia at 89.5% and Japan at 89.4%.


Cervical cancer survival rates by country

Cervical cancer is highly preventable if detected and treated before progression occurs. The main cause of cervical cancer is sexual exposure to human papilloma virus (HPV). About half of OECD countries have cervical screening programmes and around half have implemented HPV vaccination programmes.

In England, all girls aged 12 to 13 are offered the HPV vaccination, and from the 2019/20 school year the programme had been extended to all boys aged 12 to 13 years old. In 2021/22, uptake of the first dose of the vaccine was 82% and the second dose was 67%, a decrease when compared with previous years. This occurred due to disruptions in the vaccination schedule caused by school closures during the pandemic.  

International trends in five-year cervical cancer survival show more variation between countries over time than for five-year breast cancer survival. While survival has been improving in the UK, it is still one of the worst performers compared with other OECD countries, with a five-year survival of only 64% in 2010-2014. This is despite relatively high cervical cancer screening coverage. In comparison, survival in Denmark in the same period was 70% while survival in Japan reached 71%.


Colon cancer survival rates by country

Bowel cancer is the third most common cancer diagnosis worldwide. The disease has a higher prevalence in America and Europe than Asia, however in countries where people are adopting Western diets, such as Japan, the incidence of colon cancer is increasing. There are multiple screening methods available including the faecal occult blood test, colonoscopy and flexible sigmoidoscopy. Advances in diagnosis and treatment of colon cancer have led to increased survival over the last decade.

While the five-year colon cancer survival rate in the UK has been increasing over time, in 2010-2014 survival reached only 60%. This is the lowest five-year survival rate out of the 18 OECD countries that we have included. Over the same period in Australia, it was as high as 71%, while Belgium reached a five-year colon cancer survival rate of 68%.


 

About this data

The National Cancer Registration and Analysis Service (NCRAS) within NHS Digital (NHSD) data

Net survival is an estimate of the probability of survival from the cancer alone. It can be interpreted as the survival of cancer patients after considering the background mortality that the patients would have experienced if they had not had the cancer. Background mortality is derived from life tables of all-cause mortality rates in the general population. Net survival varies with age, and the age profile of cancer patients can vary with time and between geographical areas, so the estimates are age-standardised to facilitate comparison. Estimates are shown with their 95% confidence intervals. For convenience, net survival is expressed as a percentage in the range 0–100 per cent. For more information, please see the NCRAS website.

National Cancer Registration and Analysis Service data

Stage at diagnosis is based on the TNM staging system. The cancers included in the measure are prostate, breast, lung, colorectal, urological, gynaecological, melanoma and oesophago-gastric. Figures are based on rapid processing of cancer registration data and may not be complete, particularly for diagnoses in recent months. Data on the proportion of cancers diagnosed at an early stage by deprivation uses a “complete case approach” whereby only staged cancer diagnoses form the denominator from which a percentage is calculated. Since un-staged cancers have been excluded while computing the percentage, the denominator is not reflective of total cancer diagnoses. For more information, please see the Rapid Cancer Registration Dataset website and the “Uses and Limitations” tab of the CancerDatawebpage.

Caution should be taken when interpreting data for January and February 2022 as it is less complete than previous months.

OECD data

Five-year net survival is the cumulative probability that cancer patients survive their cancer for at least five years, after controlling for the risks of death from other causes. Net survival is expressed as a percentage. Net survival for patients diagnosed during 2000-2004 is based on a cohort approach, since all patients had been followed up for at least five years by the end of 2014. For patients diagnosed during 2010-2014, the period approach is used, which allows estimation of five-year survival, though five years of follow-up are not available for all patients. Cancer survival estimates are age-standardised with the International Cancer Survival Standard (ICSS) weights.

For further information please see: OECD Definitions for Health Care Quality Indicators and Health at a Glance 2019: OECD Indicators.

The OECD.Stat platform which hosted the data until June 2023 has been replaced by the OECD Data Explorer. This transfer may have affected some of the datasets, causing a delay for when numbers are updated for each country.

Comments