Guideline-consistent breast cancer surgery ranged from 13% to 82% by country

International comparisons of cancer survival have existed for years. What they cannot explain is why survival differs, because survival is an outcome and the causes sit upstream in stage at diagnosis and in what treatment was given. A study published in The Lancet on October 22 works that upstream problem directly for three cancers that affect women [s1].

The dataset

VENUSCANCER assembled anonymised individual records from 103 population-based cancer registries in 39 countries and territories, for women diagnosed with cancer of the breast, cervix or ovary during a single year of incidence between 2015 and 2018 [s1].

The registries supplied high-resolution data: stage at diagnosis, staging procedures, tumour grade, biomarkers including ER, PR and HER2 status, and the first course of each treatment modality — surgery, radiotherapy, chemotherapy, endocrine treatment or anti-HER2 therapy — with dates [s1].

In total, 275,792 records were received: 214,111 breast (77.6%), 44,468 cervix (16.1%, including in situ) and 17,213 ovary (6.2%) [s1]. Treatment was assessed against international clinical guidelines from ESMO, ASCO and NCCN, and the odds of receiving guideline-consistent treatment were compared between high-income countries and low- and middle-income countries, controlling for age and tumour subtype [s1].

Stage at diagnosis

In high-income countries, early-stage, node-negative cancers made up over 40% of breast and cervical cancers but under 20% of ovarian cancers [s1].

In low- and middle-income countries, those proportions were generally below 20% for all three cancers, with exceptions: 30% for breast cancer in Cuba, and 36% for cervix and 27% for ovary in Russia [s1].

The treatment gap

The variation in guideline consistency is wider than most readers will expect.

For surgery and radiotherapy in early-stage breast cancer, the proportion of women receiving guideline-consistent treatment ranged from 13% in Georgia to 82% in France [s1]. For chemotherapy in advanced cervical cancer, from 18% in Mongolia to 90% in Canada [s1]. For surgery plus chemotherapy in metastatic ovarian cancer, from 9% in Cuba to 53% in the USA [s1].

Some form of surgery was offered to 78% of women in high-income countries and 56% in low- and middle-income countries [s1].

An age gradient ran through both settings: women aged 70-99 had lower odds of receiving guideline-consistent initial treatment than women aged 50-69, in high-income and low- and middle-income countries alike [s1].

Time to treatment

Median time between diagnosis and treatment for early-stage cancers was under one month in several high-income countries [s1]. At the other end it reached up to four months for cervical cancer in Mongolia and for ovarian cancer in Ecuador, and up to a year for breast cancer in Mongolia [s1].

A year between diagnosis and first treatment is a different kind of failure from not being diagnosed. It means the diagnostic system worked and the treatment system did not.

The finding that complicates the simple story

Guideline-consistent initial treatment for early-stage tumours was followed more uniformly for cervical and ovarian cancer than for breast cancer [s1]. Breast cancer has the largest caseload and the most treatment options, and it is where consistency was worst.

The authors' summary is that guideline-consistent treatment has become more accessible for women diagnosed with early-stage tumours in low- and middle-income countries, but that the proportion diagnosed early remains far too low [s1]. Both halves matter: treatment access is improving, and it is being applied to a population that mostly arrives late.

What the design can and cannot support

This is a secondary analysis of registry records, not a trial. Registries record what was delivered, not why — a woman recorded as not receiving guideline-consistent treatment may have declined it, been unfit for it, had it unavailable, or been unable to reach it. Those are very different problems with very different remedies, and the data cannot separate them.

Registry coverage and data completeness vary between participating countries, and the countries that participate are not a random sample of the world. Comparisons between two named countries should be read with that in mind; the value of the dataset is in the range and the pattern, not in any single pairwise contrast.

The incidence window is 2015-2018 [s1]. Treatment patterns move, and several of these systems have changed since.

Why it was published now

An accompanying comment in the same issue set the findings alongside the CONCORD programme of global survival surveillance, framing both as the evidentiary basis for closing global gaps in cancer care for women [s2].

The study itself positions its results as real-world evidence to support the implementation and monitoring of WHO's Global Breast Cancer Initiative and its Cervical Cancer Elimination Initiative [s1]. Those programmes set targets; VENUSCANCER supplies the baseline against which the targets can be measured.

What to watch

Whether registries continue contributing high-resolution treatment data, which is considerably more demanding than incidence reporting; whether the age gradient in guideline-consistent treatment is examined as a question of clinical appropriateness rather than access; and whether later incidence years show movement in the countries at the bottom of these ranges.

This article describes registry-based observational research and is informational only. It is not medical advice.

Sources

• [s1] Global variation in patterns of care and time to initial treatment for breast, cervical, and ovarian cancer from 2015 to 2018 (VENUSCANCER): a secondary analysis of individual records for 275 792 women from 103 population-based cancer registries in 39 countries and territories, The Lancet, 2025;406(10517):2325-2348, published online 2025-10-22.
• [s2] Lessons from CONCORD and VENUSCANCER: closing global gaps in cancer care for women, The Lancet, 2025;406(10517):2298-2300, published online 2025-10-22.