Skip to main content

Together we are beating cancer

Donate now
  • Health & Medicine

Treating late-stage ovarian cancer – why does the UK do so badly?

by Henry Scowcroft | Analysis

3 October 2012

0 comments 0 comments

A patient with a nurse

On average, 19 women are diagnosed with ovarian cancer every day in the UK

One of the most important facts about cancer, which seems so obvious that repeating it again is almost banal, is that the earlier cancer is detected, the better.

Another frequently repeated ‘cancer fact’ is that UK patients tend to do worse than patients in some other developed countries.

As we’ve discussed in the past, the UK’s ‘sick man’ status in Europe is a bit of an oversimplification, and actually appears to be improving thanks to the attention successive governments have given to cancer care.

Nevertheless, study after study has highlighted the fact that the UK’s survival rates lag behind other countries. But the exact cause, or causes, of these differences has been the subject of much debate and controversy.

Is it, as some suspect, due to later diagnosis in the UK? Or is the quality of care not as good as in other countries? Do patients lack access to the latest or most appropriate forms of treatment? Are our patients generally less well, and unfit for treatments that can, for some patients, be aggressive? Or is there something biologically different about cancers among Brits that makes them harder to treat?

Answering these questions is the key to fixing the problem – and it’s likely that it’s a complex mixture of several, or all, of these factors, to a greater or lesser degree. And over recent years, much has been made of the fact that generally we need to up our national game in terms of spotting cancers earlier – something that’s hard to disagree with.

But alongside this, an international project – funded by the Department of Health and spearheaded by Cancer Research UK – has been trying to untangle the cause of the differences in national cancer survival rates.

Today this effort – known as the International Cancer Benchmarking Partnership, or ICBP – has published a detailed, in-depth comparison of survival rates for women whose ovarian cancer is diagnosed at different stages, across five countries – Australia, Norway, Canada, Denmark and the UK.

The results are both good and bad news for UK patients.

The surprising and reassuring news is that we’re as good as the other countries at spotting ovarian cancer early, and women diagnosed early don’t fare significantly worse than in other countries.

In other words, the UK’s poorer ovarian cancer survival rate, compared to other countries, isn’t because of a failure to spot the disease early.

The bad news: when the disease is spotted late, the UK seems to be worse at managing these patients than other countries, and our survival rates are lower.

And it looks as if this might be one of the main reasons why, overall, our survival stats look worse than other countries.

Let’s look at what the study found in a bit more detail, and the issues it raises.

Who carried out the study?

The ICBP is a collaborative project set up a few years ago to rigorously analyse international differences in awareness of cancer signs and symptoms, in how the disease is managed, and how this affects patients. To do this, the project relies upon accurate, up-to-date information from the countries’ healthcare systems, on a wide range of aspects of a patient’s care.

The Partnership is split into five ‘modules’, each looking at a different aspect of why our survival rates might be lagging.

Today’s new analysis, published in Gynecological Oncology, came from the ‘module one’ team (led by Cancer Research UK’s survival expert, Professor Michel Coleman) who are looking at differences in survival between different countries, and possible reasons why.

What did they measure?

The team pulled together healthcare data from women diagnosed with ovarian cancer between 2004 and 2007, in each of the five countries.  Two of the key statistics they calculated, for each country in the study, were the proportion of women diagnosed at each stage of the disease (more on this below), and the relative survival rates at each stage.

Survival rates measure the proportion of patients still alive a certain period after being diagnosed. A common measure is ‘five-year relative survival’ – so if a disease is said to have an ‘80 per cent five-year survival’, that means that, for every 100 patients diagnosed with the disease, 80 of them will survive their disease for at least five years, on average.

(We discussed survival rates in more detail in this post about the ICBP from last December)

The new ICBP study looked at the one-year survival rate for women with each of the stages. This shows directly how many patients are diagnosed at an early stage, and how many are picked up with late stage disease; and how women with different stages survive in the short term.

A cancer’s ‘stage’ is a measure of how advanced it is when doctors first diagnose it (there’s detailed information about this on our main website). But, for a variety of reasons, this isn’t routinely collected for all cancer patients. One of the main reasons that the ICBP involves the UK, Norway, Canada, Australia and Denmark is because they’re countries which have long-established, complete and reliable data.

But collecting stage data from different countries is fraught with difficulties, as there are different methods of classifying stages. One system for ovarian cancer is the FIGO system, used in the UK and widely around Europe, which defines tumours as stage I, II, III, and IV (with stage I being the earliest).

But Australia, the US and other countries use different system – SEER SS2000 – which has just three categories – ‘localised’ (corresponding to FIGO stage I), ‘regional’ (stage II) and ‘distant’ (stage III and IV combined). Since it’s possible to convert FIGO stages to SEER stages, but not the other way round, converting everything to the SEER system allowed the team to compare ovarian cancer staging across all five countries.

In all, they were able to collate information on stage and survival for just over 20,000 women in the analysis.

What did they find?

Surprisingly, given our lower overall survival, the team found that the UK actually had one of higher proportions of women diagnosed at an early stage – just under a quarter – substantially better than Denmark, Norway or Canada. You can see the data in the table below:

Country % cancers diagnosed early % cancers diagnosed at an intermediate stage % cancers diagnosed late
UK 22 14 64
Denmark 12 14 74
Australia 23 17 60
Canada 14 20 66
Norway 12 18 70

And the survival rates among early-stage patients differed between the countries, with Denmark doing particularly badly compared to the rest.

However, when you look at the one-year survival for late-stage disease, the UK’s figures (61 per cent) were significantly lower than the rest, with Canada leading the way:

Country 1yr survival (early stage) 1yr survival (medium stage) 1yr survival
(late stage)
UK 98 93 61 69
Denmark 92 96 67 73
Australia 94 84 65 75
Canada 98 94 76 74
Norway 97 95 66 74

This strongly suggests that late detection of ovarian cancer is not the reason we lag behind other countries. Instead it seems that our treatment for patients with late-stage disease may be dragging down the overall survival rate for ovarian cancer in this country.

But why?

Quality of, and access to, care

One obvious possibility, according to the authors, is that the UK’s women aren’t receiving the best treatment. There could be a variety of reasons for this:

  • The standard of medical care in this country could be lower than elsewhere.
  • Doctors may not be able to access the latest treatment options – be they surgery, chemo or radiotherapy – for their patients.

But there are three other possibilities:

  • British patients may be more likely to have other conditions, like heart disease, diabetes, etc (technically referred to as ‘co-morbidities’) so they may be less able to tolerate aggressive treatments than patients in other countries.
  • There may be differences in the types of ovarian cancer that may be more common in the UK (there are two main forms of ovarian cancers: serous/high grade, which are often aggressive and fast-growing, and so only detectable by the time the cancer has spread; and clear-cell/mucinous, which are often less aggressive and hence diagnosed at an earlier stage).
  • There may be problems with the data – for example, the wrong stage being recorded at diagnosis, or other types of cancer being recorded as ovarian cancer.

On this last point, the study was designed in such a way that its authors don’t think that what they’ve found is due to flaws in the data or the design of the analysis.

But ruling out any of the other possibilities is impossible at this point, as we don’t yet have detailed information on what treatments patients received, how long they were treated for and what other conditions they had. We urgently need further studies to focus on these issues, and to try to pin down why these differences are occurring. Only then can we take action to fix it. Thankfully the ICBP is planning to focus on this over the coming months and years.

And the ICBP is also working on similar analyses for breast, bowel and lung cancers – it will be interesting to see whether the worrying issues raised here, of poor survival among people treated for late stage disease, are confined to ovarian cancer, or are more widespread.

In praise of data

Over the last few years, there’s been a seismic shift in how healthcare services collect and manage the sort of information that made this study possible – and the UK has been one of the world leaders in this change. And it makes a difference. For example, we’ve blogged recently about how lung cancer surgery rates appear to be improving thanks to data identifying the worst performers.

But it hasn’t gone far enough. Despite what this study has achieved, it also found that a shocking 32 per cent of UK ovarian cancer patients had no stage data recorded in the cancer registries, compared with just below 4 per cent in Norway. Until stage and treatment data are routinely stored for all cancer patients, in a consistent, accessible and accurate manner, we simply won’t know enough about how to improve the way healthcare services for cancer patients are structured and managed.

And given the dramatic changes already underway across the NHS in England, it’s vital that the progress made so far is protected and built upon.

It’s a cliché, but data can actually help save lives. We must do everything we can to make sure it continues to do so.



  • Maringe, C. et al (2012). Stage at diagnosis and ovarian cancer survival: Evidence from the International Cancer Benchmarking Partnership, Gynecologic Oncology, 127 (1) 82. DOI: 10.1016/j.ygyno.2012.06.033