Combining simple blood tests to identify primary care patients with unexpected weight loss for cancer investigation: Clinical risk score development, internal validation, and net benefit analysis

A review of methods for the analysis of diagnostic tests performed in sequence

BACKGROUND

Many clinical pathways for the diagnosis of disease are based on diagnostic tests that are performed in sequence. The performance of the full diagnostic sequence is dictated by the diagnostic performance of each test in the sequence as well as the conditional dependence between them, given true disease status. Resulting estimates of performance, such as the sensitivity and specificity of the test sequence, are key parameters in health-economic evaluations. We conducted a methodological review of statistical methods for assessing the performance of diagnostic tests performed in sequence, with the aim of guiding data analysts towards classes of methods that may be suitable given the design and objectives of the testing sequence.

METHODS

We searched PubMed, Scopus and Web of Science for relevant papers describing methodology for analysing sequences of diagnostic tests. Papers were classified by the characteristics of the method used, and these were used to group methods into themes. We illustrate some of the methods using data from a cohort study of repeat faecal immunochemical testing for colorectal cancer in symptomatic patients, to highlight the importance of allowing for conditional dependence in test sequences and adjustment for an imperfect reference standard.

RESULTS

Five overall themes were identified, detailing methods for combining multiple tests in sequence, estimating conditional dependence, analysing sequences of diagnostic tests used for risk assessment, analysing test sequences in conjunction with an imperfect or incomplete reference standard, and meta-analysis of test sequences.

CONCLUSIONS

This methodological review can be used to help researchers identify suitable analytic methods for studies that use diagnostic tests performed in sequence.

Trends in primary care blood tests prior to lung and colorectal cancer diagnosis-A retrospective cohort study using linked Australian data

INTRODUCTION

Abnormal results in common blood tests may occur several months before lung cancer (LC) and colorectal cancer (CRC) diagnosis. Identifying early blood markers of cancer and distinct blood test signatures could support earlier diagnosis in general practice.

METHODS

Using linked Australian primary care and hospital cancer registry data, we conducted a cohort study of 855 LC and 399 CRC patients diagnosed between 2001 and 2021. Requests and results from general practice blood tests (six acute phase reactants [APR] and six red blood cell indices [RBCI]) were examined in the 2 years before cancer diagnosis. Poisson regression models were used to estimate monthly incidence rates and examine pre-diagnostic trends in blood test use and abnormal results prior to cancer diagnosis, comparing patterns in LC and CRC patients.

RESULTS

General practice blood test requests increase from 7 months before CRC and 6 months before LC diagnosis. Abnormalities in many APR and RBCI tests increase several months before cancer diagnosis, often occur prior to or in the absence of anaemia (in 51% of CRC and 81% of LC patients with abnormalities), and are different in LC and CRC patients.

CONCLUSIONS

This study demonstrates an increase in diagnostic activity in Australian general practice several months before LC and CRC diagnosis, indicating potential opportunities for earlier diagnosis. It identifies blood test abnormalities and distinct signatures that are early markers of LC and CRC. If combined with other pre-diagnostic information, these blood tests have potential to support GPs in prioritising patients for cancer investigation of different sites to expedite diagnosis.

Predictive value of abnormal blood tests for detecting cancer in primary care patients with nonspecific abdominal symptoms: A population-based cohort study of 477,870 patients in England

BACKGROUND

Identifying patients presenting with nonspecific abdominal symptoms who have underlying cancer is a challenge. Common blood tests are widely used to investigate these symptoms in primary care, but their predictive value for detecting cancer in this context is unknown. We quantify the predictive value of 19 abnormal blood test results for detecting underlying cancer in patients presenting with 2 nonspecific abdominal symptoms.

METHODS AND FINDINGS

Using data from the UK Clinical Practice Research Datalink (CPRD) linked to the National Cancer Registry, Hospital Episode Statistics and Index of Multiple Deprivation, we conducted a population-based cohort study of patients aged ≥30 presenting to English general practice with abdominal pain or bloating between January 2007 and October 2016. Positive and negative predictive values (PPV and NPV), sensitivity, and specificity for cancer diagnosis (overall and by cancer site) were calculated for 19 abnormal blood test results co-occurring in primary care within 3 months of abdominal pain or bloating presentations. A total of 9,427/425,549 (2.2%) patients with abdominal pain and 1,148/52,321 (2.2%) with abdominal bloating were diagnosed with cancer within 12 months post-presentation. For both symptoms, in both males and females aged ≥60, the PPV for cancer exceeded the 3% risk threshold used by the UK National Institute for Health and Care Excellence for recommending urgent specialist cancer referral. Concurrent blood tests were performed in two thirds of all patients (64% with abdominal pain and 70% with bloating). In patients aged 30 to 59, several blood abnormalities updated a patient's cancer risk to above the 3% threshold: For example, in females aged 50 to 59 with abdominal bloating, pre-blood test cancer risk of 1.6% increased to: 10% with raised ferritin, 9% with low albumin, 8% with raised platelets, 6% with raised inflammatory markers, and 4% with anaemia. Compared to risk assessment solely based on presenting symptom, age and sex, for every 1,000 patients with abdominal bloating, assessment incorporating information from blood test results would result in 63 additional urgent suspected cancer referrals and would identify 3 extra cancer patients through this route (a 16% relative increase in cancer diagnosis yield). Study limitations include reliance on completeness of coding of symptoms in primary care records and possible variation in PPVs if extrapolated to healthcare settings with higher or lower rates of blood test use.

CONCLUSIONS

In patients consulting with nonspecific abdominal symptoms, the assessment of cancer risk based on symptoms, age and sex alone can be substantially enhanced by considering additional information from common blood test results. Male and female patients aged ≥60 presenting to primary care with abdominal pain or bloating warrant consideration for urgent cancer referral or investigation. Further cancer assessment should also be considered in patients aged 30 to 59 with concurrent blood test abnormalities. This approach can detect additional patients with underlying cancer through expedited referral routes and can guide decisions on specialist referrals and investigation strategies for different cancer sites.

Nonspecific Signs and/or Symptoms of Cancer: A Retrospective, Observational Analysis from a Secondary Care, US Community Oncology Dataset

To help determine the unmet need for improved diagnostic tools to evaluate patients with nonspecific signs and/or symptoms (NSSS) and suspicion of cancer, we examined patient characteristics, diagnostic journey, and cancer incidence of patients with NSSS within The US Oncology Network (The Network), a secondary care community oncology setting. This retrospective, observational cohort study included patients aged ≥40 years with ≥1 NSSS in their problem list at their first visit within The Network (the index date) between 1 January 2016 and 31 December 2020. Patients were followed longitudinally with electronic health record data for initial cancer diagnosis, new noncancer diagnosis, death, end of study observation period, or 12 months, whichever occurred first. Of 103,984 patients eligible for inclusion, 96,722 presented with only 1 NSSS at index date; 6537/103,984 (6.3%) were diagnosed with 1 primary cancer within 12 months after the index date; 3825/6537 (58.5%) with hematologic malignancy, and 2712/6537 (41.5%) with solid tumor. Among patients diagnosed with cancer (n = 6774), the median time to cancer diagnosis after their first visit within The Network was 5.13 weeks. This study provides a real-world perspective on cancer incidence in patients with NSSS referred to a secondary care setting and highlights the unmet need for improved diagnostic tools to improve cancer outcomes.

Cancer Diagnoses After Recent Weight Loss

Importance

Weight loss is common in primary care. Among individuals with recent weight loss, the rates of cancer during the subsequent 12 months are unclear compared with those without recent weight loss.

Objective

To determine the rates of subsequent cancer diagnoses over 12 months among health professionals with weight loss during the prior 2 years compared with those without recent weight loss.

Design, Setting, and Participants

Prospective cohort analysis of females aged 40 years or older from the Nurses' Health Study who were followed up from June 1978 until June 30, 2016, and males aged 40 years or older from the Health Professionals Follow-Up Study who were followed up from January 1988 until January 31, 2016.

Exposure

Recent weight change was calculated from the participant weights that were reported biennially. The intentionality of weight loss was categorized as high if both physical activity and diet quality increased, medium if only 1 increased, and low if neither increased.

Main Outcome and Measures

Rates of cancer diagnosis during the 12 months after weight loss.

Results

Among 157 474 participants (median age, 62 years [IQR, 54-70 years]; 111 912 were female [71.1%]; there were 2631 participants [1.7%] who self-identified as Asian, Native American, or Native Hawaiian; 2678 Black participants [1.7%]; and 149 903 White participants [95.2%]) and during 1.64 million person-years of follow-up, 15 809 incident cancer cases were identified (incident rate, 964 cases/100 000 person-years). During the 12 months after reported weight change, there were 1362 cancer cases/100 000 person-years among all participants with recent weight loss of greater than 10.0% of body weight compared with 869 cancer cases/100 000 person-years among those without recent weight loss (between-group difference, 493 cases/100 000 person-years [95% CI, 391-594 cases/100 000 person-years]; P < .001). Among participants categorized with low intentionality for weight loss, there were 2687 cancer cases/100 000 person-years for those with weight loss of greater than 10.0% of body weight compared with 1220 cancer cases/100 000 person-years for those without recent weight loss (between-group difference, 1467 cases/100 000 person-years [95% CI, 799-2135 cases/100 000 person-years]; P < .001). Cancer of the upper gastrointestinal tract (cancer of the esophagus, stomach, liver, biliary tract, or pancreas) was particularly common among participants with recent weight loss; there were 173 cancer cases/100 000 person-years for those with weight loss of greater than 10.0% of body weight compared with 36 cancer cases/100 000 person-years for those without recent weight loss (between-group difference, 137 cases/100 000 person-years [95% CI, 101-172 cases/100 000 person-years]; P < .001).

Conclusions and Relevance

Health professionals with weight loss within the prior 2 years had a significantly higher risk of cancer during the subsequent 12 months compared with those without recent weight loss. Cancer of the upper gastrointestinal tract was particularly common among participants with recent weight loss compared with those without recent weight loss.

Multi-cancer early detection test in symptomatic patients referred for cancer investigation in England and Wales (SYMPLIFY): a large-scale, observational cohort study

BACKGROUND

Analysis of circulating tumour DNA could stratify cancer risk in symptomatic patients. We aimed to evaluate the performance of a methylation-based multicancer early detection (MCED) diagnostic test in symptomatic patients referred from primary care.

METHODS

We did a multicentre, prospective, observational study at National Health Service (NHS) hospital sites in England and Wales. Participants aged 18 or older referred with non-specific symptoms or symptoms potentially due to gynaecological, lung, or upper or lower gastrointestinal cancers were included and gave a blood sample when they attended for urgent investigation. Participants were excluded if they had a history of or had received treatment for an invasive or haematological malignancy diagnosed within the preceding 3 years, were taking cytotoxic or demethylating agents that might interfere with the test, or had participated in another study of a GRAIL MCED test. Patients were followed until diagnostic resolution or up to 9 months. Cell-free DNA was isolated and the MCED test performed blinded to the clinical outcome. MCED predictions were compared with the diagnosis obtained by standard care to establish the primary outcomes of overall positive and negative predictive value, sensitivity, and specificity. Outcomes were assessed in participants with a valid MCED test result and diagnostic resolution. SYMPLIFY is registered with ISRCTN (ISRCTN10226380) and has completed follow-up at all sites.

FINDINGS

6238 participants were recruited between July 7 and Nov 30, 2021, across 44 hospital sites. 387 were excluded due to staff being unable to draw blood, sample errors, participant withdrawal, or identification of ineligibility after enrolment. Of 5851 clinically evaluable participants, 376 had no MCED test result and 14 had no information as to final diagnosis, resulting in 5461 included in the final cohort for analysis with an evaluable MCED test result and diagnostic outcome (368 [6·7%] with a cancer diagnosis and 5093 [93·3%] without a cancer diagnosis). The median age of participants was 61·9 years (IQR 53·4-73·0), 3609 (66·1%) were female and 1852 (33·9%) were male. The MCED test detected a cancer signal in 323 cases, in whom 244 cancer was diagnosed, yielding a positive predictive value of 75·5% (95% CI 70·5-80·1), negative predictive value of 97·6% (97·1-98·0), sensitivity of 66·3% (61·2-71·1), and specificity of 98·4% (98·1-98·8). Sensitivity increased with increasing age and cancer stage, from 24·2% (95% CI 16·0-34·1) in stage I to 95·3% (88·5-98·7) in stage IV. For cases in which a cancer signal was detected among patients with cancer, the MCED test's prediction of the site of origin was accurate in 85·2% (95% CI 79·8-89·3) of cases. Sensitivity 80·4% (95% CI 66·1-90·6) and negative predictive value 99·1% (98·2-99·6) were highest for patients with symptoms mandating investigation for upper gastrointestinal cancer.

INTERPRETATION

This first large-scale prospective evaluation of an MCED diagnostic test in a symptomatic population demonstrates the feasibility of using an MCED test to assist clinicians with decisions regarding urgency and route of referral from primary care. Our data provide the basis for a prospective, interventional study in patients presenting to primary care with non-specific signs and symptoms.

FUNDING

GRAIL Bio UK.

Symptoms and signs of lung cancer prior to diagnosis: case-control study using electronic health records from ambulatory care within a large US-based tertiary care centre

OBJECTIVE

Lung cancer is the most common cause of cancer-related death in the USA. While most patients are diagnosed following symptomatic presentation, no studies have compared symptoms and physical examination signs at or prior to diagnosis from electronic health records (EHRs) in the USA. We aimed to identify symptoms and signs in patients prior to diagnosis in EHR data.

DESIGN

Case-control study.

SETTING

Ambulatory care clinics at a large tertiary care academic health centre in the USA.

PARTICIPANTS, OUTCOMES

We studied 698 primary lung cancer cases in adults diagnosed between 1 January 2012 and 31 December 2019, and 6841 controls matched by age, sex, smoking status and type of clinic. Coded and free-text data from the EHR were extracted from 2 years prior to diagnosis date for cases and index date for controls. Univariate and multivariable conditional logistic regression were used to identify symptoms and signs associated with lung cancer at time of diagnosis, and 1, 3, 6 and 12 months before the diagnosis/index dates.

RESULTS

Eleven symptoms and signs recorded during the study period were associated with a significantly higher chance of being a lung cancer case in multivariable analyses. Of these, seven were significantly associated with lung cancer 6 months prior to diagnosis: haemoptysis (OR 3.2, 95% CI 1.9 to 5.3), cough (OR 3.1, 95% CI 2.4 to 4.0), chest crackles or wheeze (OR 3.1, 95% CI 2.3 to 4.1), bone pain (OR 2.7, 95% CI 2.1 to 3.6), back pain (OR 2.5, 95% CI 1.9 to 3.2), weight loss (OR 2.1, 95% CI 1.5 to 2.8) and fatigue (OR 1.6, 95% CI 1.3 to 2.1).

CONCLUSIONS

Patients diagnosed with lung cancer appear to have symptoms and signs recorded in the EHR that distinguish them from similar matched patients in ambulatory care, often 6 months or more before diagnosis. These findings suggest opportunities to improve the diagnostic process for lung cancer.

Development and validation of a routine blood parameters-based model for screening the occurrence of retinal detachment in high myopia in the context of PPPM

Background/aims

Timely detection and treatment of retinal detachment (RD) could effectively save vision and reduce the risk of progressing visual field defects. High myopia (HM) is known to be associated with an increased risk of RD. Evidently, it should be clearly discriminated the individuals with high or low risk of RD in patients with HM. By using multi-parametric analysis, risk assessment, and other techniques, it is crucial to create cutting-edge screening programs that may be utilized to improve population eye health and develop person-specific, cost-effective preventative, and targeted therapeutic measures. Therefore, we propose a novel, routine blood parameters-based prediction model as a screening program to help distinguish who should offer detailed ophthalmic examinations for RD diagnosis, prevent visual field defect progression, and provide personalized, serial monitoring in the context of predictive, preventive, and personalized medicine (PPPM/3 PM).

Methods

This population-based study included 20,870 subjects (HM = 19,284, HMRD = 1586) who underwent detailed routine blood tests and ophthalmic evaluations. HMRD cases and HM controls were matched using a nested case-control design. Then, the HMRD cases and HM controls were randomly assigned to the discovery cohort, validation cohort 1, and validation cohort 2 maintaining a 6:2:2 ratio, and other subjects were assigned to the HM validation cohort. Receiver operating characteristic curve analysis was performed to select feature indexes. Feature indexes were integrated into seven algorithm models, and an optimal model was selected based on the highest area under the curve (AUC) and accuracy.

Results

Six feature indexes were selected: lymphocyte, basophil, mean platelet volume, platelet distribution width, neutrophil-to-lymphocyte ratio, and lymphocyte-to-monocyte ratio. Among the algorithm models, the algorithm of conditional probability (ACP) showed the best performance achieving an AUC of 0.79, a diagnostic accuracy of 0.72, a sensitivity of 0.71, and a specificity of 0.74 in the discovery cohort. A good performance of the ACP model was also observed in the validation cohort 1 (AUC = 0.81, accuracy = 0.72, sensitivity = 0.71, specificity = 0.73) and validation cohort 2 (AUC = 0.77, accuracy = 0.71, sensitivity = 0.70, specificity = 0.72). In addition, ACP model calibration was found to be good across three cohorts. In the HM validation cohort, the ACP model achieved a diagnostic accuracy of 0.81 for negative classification.

Conclusion

We have developed a routine blood parameters-based model with an ACP algorithm that could potentially be applied in the clinic with a PPPM approach for serial monitoring and predicting the occurrence of RD in HM and can facilitate the prevention of HM progression to RD. According to the current study, routine blood measures are essential in patient risk classification, predictive diagnosis, and targeted therapy. Therefore, for high-risk RD persons, novel screening programs and prompt treatment plans are essential to enhance individual outcomes and healthcare offered to the community with HM.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13167-023-00319-3.

Progress and priorities in reducing the time to cancer diagnosis

Key developments in early diagnosis research and policy since the publication of the highly cited BJC review "Is increased time to diagnosis and treatment associated with poorer outcomes?" by Neal et al. in 2015 are summarised. Progress achieved since 2015 is described and priorities for further research identified.

BLOod Test Trend for cancEr Detection (BLOTTED): protocol for an observational and prediction model development study using English primary care electronic health record data

BACKGROUND

Simple blood tests can play an important role in identifying patients for cancer investigation. The current evidence base is limited almost entirely to tests used in isolation. However, recent evidence suggests combining multiple types of blood tests and investigating trends in blood test results over time could be more useful to select patients for further cancer investigation. Such trends could increase cancer yield and reduce unnecessary referrals. We aim to explore whether trends in blood test results are more useful than symptoms or single blood test results in selecting primary care patients for cancer investigation. We aim to develop clinical prediction models that incorporate trends in blood tests to identify the risk of cancer.

METHODS

Primary care electronic health record data from the English Clinical Practice Research Datalink Aurum primary care database will be accessed and linked to cancer registrations and secondary care datasets. Using a cohort study design, we will describe patterns in blood testing (aim 1) and explore associations between covariates and trends in blood tests with cancer using mixed-effects, Cox, and dynamic models (aim 2). To build the predictive models for the risk of cancer, we will use dynamic risk modelling (such as multivariate joint modelling) and machine learning, incorporating simultaneous trends in multiple blood tests, together with other covariates (aim 3). Model performance will be assessed using various performance measures, including c-statistic and calibration plots.

DISCUSSION

These models will form decision rules to help general practitioners find patients who need a referral for further investigation of cancer. This could increase cancer yield, reduce unnecessary referrals, and give more patients the opportunity for treatment and improved outcomes.

Serious disease risk among patients with unexpected weight loss: a matched cohort of over 70 000 primary care presentations

BACKGROUND

Unexpected weight loss (UWL) in patients consulting in primary care presents dilemmas for management because of the broad differential diagnoses associated with UWL. Research on the risks of serious disease among patients with UWL to date has largely taken place in secondary care, limiting generalizability to primary care patients. In this study, we use a large matched cohort study to estimate the risks of 12 serious diseases among patients presenting to primary care with UWL where this was recorded, stratified by age and sex, in order to inform a rational clinical approach to patients presenting with UWL.

METHODS

This was a retrospective matched cohort study using electronic health records (EHRs) from the UK Clinical Practice Research Datalink (CPRD). Each patient with UWL (ascertained from EHR coding) was matched to five patients without UWL and followed until the earliest of a diagnosis of the serious disease, date of death, exit from the CPRD database, or end of the study. Observed absolute risks of the 12 serious diseases were estimated as probabilities, and hazard ratios (HRs) were estimated with Cox proportional hazards models.

RESULTS

Between 2000 and 2012, 70 193 patients in CPRD had at least one record of UWL and were matched with 295 579 patients without UWL. Patients with UWL had significantly higher risk of nearly all serious diseases examined compared with patients without. HRs ranged from 1.43 for congestive heart failure [95% confidence interval (CI): 1.27-1.62] to 9.70 for malabsorption (95% CI: 6.81-13.82). The absolute risks of any given serious disease were relatively low (<6% after 1 year). The magnitude and rank order of absolute risks varied by age and sex. Depression was the most common diagnosis among women aged <80 with UWL (3.74% of women aged <60 and 2.46% of women aged 60-79), whereas diabetes was the most common in men <60 with UWL (2.96%) and cancer was the most common in men aged 60 and over with UWL (3.79% of men aged 60-70 and 5.28% of men aged ≥80).

CONCLUSIONS

This analysis provides new evidence to patients and clinicians about the risks of serious disease among patients presenting with UWL in primary care. Depending on age and sex, the results suggest that workup for UWL should include screening for diabetes, thyroid dysfunction, depression, and dementia. If performed in a timely manner, this workup could be used to triage patients eligible for cancer pathway referral.

Home-built environment interventions and inflammation biomarkers: a systematic review and meta-analysis protocol

BACKGROUND

Inflammation control is a fundamental part of chronic care in patients with a history of cancer and comorbidity. As the risk-benefit profile of anti-inflammatory drugs is unclear in survivors of cancer, GPs and patients could benefit from alternative non-pharmacological treatment options for dysregulated inflammation. There is a potential for home-built environment (H-BE) interventions to modulate inflammation; however, discrepancies exist between studies.

AIM

To evaluate the effectiveness of H-BE interventions on cancer-associated inflammation biomarkers.

DESIGN & SETTING

A systematic review and meta-analysis of randomised and non-randomised trials in community-dwelling adults.

METHOD

PubMed and MEDLINE, Embase, Web of Science, and Google Scholar will be searched for clinical trials published in January 2000 onwards. The study will include H-BE interventions modifying air quality, thermal comfort, non-ionising radiation, noise, nature, and water. No restrictions to study population will be applied to allow deriving expectations for effects of the interventions in cancer survivors from available source populations. Outcome measures will be inflammatory biomarkers clinically and physiologically relevant to cancer. The first reviewer will independently screen articles together with GPs and extract data that will be verified by a second reviewer. The quality of studies will be assessed using the Cochrane risk-of-bias tools. Depending on the clinical and methodological homogeneity of populations, interventions, and outcomes, a meta-analysis will be conducted using random-effects models.

CONCLUSION

Findings will determine the effectiveness of H-BE interventions on inflammatory parameters, guide future directions for its provision in community-dwelling survivors of cancer and support GPs with safer anti-inflammatory treatment options in high-risk patients for clinical complications.

Measured weight loss as a precursor to cancer diagnosis: retrospective cohort analysis of 43 302 primary care patients

BACKGROUND

Unexpected weight loss is a presenting feature of cancer in primary care. Data from primary care are lacking to quantify how much weight loss over what period should trigger further investigation for cancer. This research aimed to quantify cancer diagnosis rates associated with measured weight change in people attending primary care.

METHODS

Retrospective cohort study of primary care electronic health records data linked to the Surveillance, Epidemiology, and End Results cancer registry (Integrated healthcare delivery system in Washington State, United States). Multivariable Cox regression incorporating time varying covariates using splines to model non-linear associations (age, percentage weight change, and weight change interval). Fifty thousand randomly selected patients aged 40 years and over followed for up to 9 years (1 January 2006 to 31 December 2014). Outcome measures are hazard ratios (95% confidence intervals) to quantify the association between percentage weight change and cancer diagnosis for all cancers combined, individual cancer sites and stages; percentage risk of cancer diagnosis within 6 months of the end of each weight change episode; and the positive predictive value for cancer diagnosis.

RESULTS

There were 43 302 included in the analysis after exclusions. Over 287 858 patient-years of follow-up, including 24 272 (56.1%) females, 23 980 (55.4%) aged 40 to 59 years, 15 113 (34.9%) 60 to 79 years, and 4209 (9.7%) aged 80 years and over. Adjusted hazard ratios (95% confidence interval) for cancer diagnosis in a 60 years old ranged from 1.04 (1.02 to 1.05, P < 0.001) for 1% weight loss to 1.44 (1.23 to 1.68, P < 0.001) for 10%. An independent linear association was observed between percentage weight loss and increasing cancer risk. The absolute risk of cancer diagnosis increased with increasing age (up to 85 years) and as the weight change measurement interval decreased (<1 year). The positive predictive value for a cancer diagnosis within 1 year of ≥5% measured weight loss in a 60 to 69 years old was 3.41% (1.57% to 6.37%) in men and 3.47% (1.68% to 6.29%) in women. The risk of cancer diagnosis was significantly increased for pancreatic, myeloma, gastro-oesophageal, colorectal, breast, stage II and IV cancers.

CONCLUSIONS

Weight loss is a sign of undiagnosed cancer regardless of the interval over which it occurs. Guidelines should resist giving an arbitrary cut-off for the interval of weight loss and focus on the percentage of weight loss and the patient's age. Future studies should focus on the association between diagnostic evaluation of weight change and risk of cancer mortality.

Using artificial intelligence in a primary care setting to identify patients at risk for cancer: a risk prediction model based on routine laboratory tests

OBJECTIVES

To evaluate the ability of an artificial intelligence (AI) model to predict the risk of cancer in patients referred from primary care based on routine blood tests. Results obtained with the AI model are compared to results based on logistic regression (LR).

METHODS

An analytical profile consisting of 25 predefined routine laboratory blood tests was introduced to general practitioners (GPs) to be used for patients with non-specific symptoms, as an additional tool to identify individuals at increased risk of cancer. Consecutive analytical profiles ordered by GPs from November 29th 2011 until March 1st 2020 were included. AI and LR analysis were performed on data from 6,592 analytical profiles for their ability to detect cancer. Cohort I for model development included 5,224 analytical profiles ordered by GP's from November 29th 2011 until the December 31st 2018, while 1,368 analytical profiles included from January 1st 2019 until March 1st 2020 constituted the "out of time" validation test Cohort II. The main outcome measure was a cancer diagnosis within 90 days.

RESULTS

The AI model based on routine laboratory blood tests can provide an easy-to use risk score to predict cancer within 90 days. Results obtained with the AI model were comparable to results from the LR model. In the internal validation Cohort IB, the AI model provided slightly better results than the LR analysis both in terms of the area under the receiver operating characteristics curve (AUC) and PPV, sensitivity/specificity while in the "out of time" validation test Cohort II, the obtained results were comparable.

CONCLUSIONS

The AI risk score may be a valuable tool in the clinical decision-making. The score should be further validated to determine its applicability in other populations.

Cancer Special Issue: Early detection and minimal residual disease

Beryne Odeny discusses PLOS Medicine's Special Issue on early cancer detection and minimal residual disease.