When does unexpected weight loss warrant cancer investigation?

Researching whether there is an association between weight loss and cancer may seem like investigating the obvious. But with so many possible causes, which patient with weight loss warrants invasive cancer investigation?

After collating the available data on the topic, we received requests from GPs for a list of additional symptoms signs and investigations to use when assessing whether a person with unexpected weight loss should undergo cancer investigation. [1] Lists do exist, imprinted in the minds of GPs, underpinning their clinical approach, developed through experience to become their diagnostic rule of thumb. We could find no such list published nor appraise how accurate they were.  

Guidelines may miss matters of clinical salience without research to document them. From guidelines grow clinical pathways (in many countries) around which the machinery of cancer diagnosis is aligned. [2] Routes into the diagnostic system may therefore be missing or convoluted for important presentations if the research has not been done. For these patients, a cancer diagnosis may be delayed, at a later stage, with poorer outcomes.  

Unexpected weight loss is, of course, associated with cancer. We learnt this at medical school. It is associated with a broad range of cancers in the three to six months after a first presentation to primary care. [3] But only one to two people will have cancer diagnosed out of one hundred people presenting with it. Investigating each new presentation of weight loss could cause preventable harm from the three o’s: over-investigation, overdiagnosis, and over-treatment. [4] In our recent research paper, we provide evidence for which symptoms, signs, and blood test results increase the risk of cancer in people with unexpected weight lossWe present evidence based referral criteria for clinicians and guideline developers.  

Simple-to-use symptom-derived cancer referral criteria are effective at ruling-in people for cancer investigation. Prioritising patients for investigation remains the priority of health policy but provides only one piece of the puzzle. A key attribute of primary care is reassuring patients they do not require investigation. Multivariable risk scores may be better at ruling people out of investigation by combining multiple negative findings. There remains poor uptake of cancer risk scores in clinical practice. [5] Primary care clinicians select which symptomatic patients to refer by incorporating multiple measurable and unmeasurable pieces of information. The intensity, persistence, and duration of weight loss is assessed in the context of previous occurrences, comorbid conditions, and changes in diet and daily exercise (enforced by covid-19 lockdown). Many of these rich contextual factors are poorly recorded in health records data used to derive risk scores. [6] As clinicians, we know this. 

Covid-19 has sharpened our focus on maximising diagnostic system efficiency. [7] Why review in person if we can adequately review remotely? Why review prior to testing if testing is inevitable? Can we use triage tests more effectively in primary care? Are additional appointments necessary if all tests are performed together? Multidisciplinary Diagnostic Centres (MDC) were piloted across the U.K. to answer these sorts of questions. [8] The roll-out of their offspring, NHS Rapid Diagnostic Centres, has stalled during the pandemic. Early MDC evaluations remind us that the investigation of non-specific symptoms such as weight loss does not end with a single cancer (multiple cancers are associated with weight loss), or with cancer (three times as many other serious diseases are diagnosed than cancer). [9] While rebuilding and reopening the health system we should take forward emergent partnerships for healthcare delivery and data science that instil diagnostic safety and efficiency. This could benefit all patients, including those with symptoms that could be caused by cancer.  

Brian Nicholson is a GP and an NIHR Academic Clinical Lecturer based in Oxford. His research aims to improve the diagnostic process for patients attending their GP. 

Competing interests: See research paper for full declaration


  1. Nicholson BD, Aveyard P, Hamilton W, Hobbs FDR. When should unexpected weight loss warrant further investigation to exclude cancer? BMJ  2019;366(I5271).
  2. Nicholson BD, Perera R, Thompson MJ. The elusive diagnosis of cancer: testing times. The British journal of general practice : the journal of the Royal College of General Practitioners. 2018;68(676):510-1.
  3. Nicholson BD, Hamilton W, Koshiaris C, Oke JL, Hobbs FDR, Aveyard P. The association between unexpected weight loss and cancer diagnosis in primary care: a matched cohort analysis of 65,000 presentations. Br J Cancer. 2020;122(12):1848-56.
  4. Kale MS, Korenstein D. Overdiagnosis in primary care: framing the problem and finding solutions. Bmj. 2018;362:k2820.
  5. Price S, Spencer A, Medina-Lara A, Hamilton W. Availability and use of cancer decision-support tools: a cross-sectional survey of UK primary care. The British journal of general practice : the journal of the Royal College of General Practitioners. 2019;69(684):e437-e43.
  6. Lyratzopoulos G, Abel GA. Assessing patients at risk of symptomatic-but-as-yet-undiagnosed cancer in primary care using information from patient records. Br J Cancer. 2020;122(12):1729-31.
  7. Hamilton W. Cancer diagnostic delay in the COVID-19 era: what happens next? Lancet Oncol. 2020;21(8):1000-2.
  8. Nicholson BD, Oke J, Friedemann Smith C, Phillips JA, Lee J, Abel L, et al. The Suspected CANcer (SCAN) pathway: protocol for evaluating a new standard of care for patients with non-specific symptoms of cancer. BMJ open. 2018;8(1):e018168.
  9. Chapman D, Poirier V, Vulkan D, Fitzgerald K, Rubin G, Hamilton W, et al. First results from five multidisciplinary diagnostic centre (MDC) projects for non-specific but concerning symptoms, possibly indicative of cancer. Br J Cancer. 2020.