Blood markers for cancer

Exploring ethnic differences in the distribution of blood test results in healthy adult populations to inform earlier cancer detection: a systematic review

BACKGROUND

In primary care, health professionals use blood tests to investigate nonspecific presentations to inform referral decisions. Reference ranges for the commonly used blood tests in western countries were developed in predominately White populations, and so may perform differently when applied to non-White populations. Knowledge of ethnic variation in blood test results in healthy/general populations could help address ethnic inequalities in cancer referral for diagnosis and outcomes.

OBJECTIVE

This systematic review explored evidence of ethnic differences in the distribution of selected blood test results among healthy/general populations to inform future research aimed at addressing inequalities in cancer diagnosis.

METHODS

We searched PubMed and EMBASE to identify studies reporting measures of haemoglobin, MCV, calcium, albumin, platelet count, and CRP in nondiseased adults from at least 2 different ethnic groups. Two reviewers independently screened studies, completed data extraction and quality assessment using an adapted Newcastle-Ottawa scale. Participants were stratified into White, Black, Asian, Mixed, and Other groups. Data were synthesised narratively and meta-analyses were conducted where possible.

RESULTS

A total of 47 papers were included. Black men and women have lower average values of haemoglobin, MCV, and albumin, and higher average values of CRP relative to their White counterparts. Additionally, Black men have lower average haemoglobin than Asian men, whereas Asian women have lower average CRP values when compared with White women.

CONCLUSIONS

There is evidence of ethnic differences in average values of haemoglobin, MCV, CRP, and albumin in healthy/general populations. Further research is needed to explore the reasons for these differences. Systematic review registration: CRD42021274580.

Advancing Gastrointestinal Cancer Diagnostics: A Systematic Review and Meta-Analysis of Circulating microRNA-1246 as a Non-invasive Biomarker

Background: Despite numerous reports on the alterations of microRNA-1246 (miR-1246) expression level in digestive system cancers, its role in gastrointestinal cancers (GICs) remains unclear. This meta-analysis aimed to assess the diagnostic potential of circulating miR-1246 in GICs.Methods: Meta-disc V.1.4 and Comprehensive Meta-Analysis V.3.7 software were used to calculate pooled sensitivity, specificity, likelihood ratios, diagnostic odds ratio, AUC, Q*index, and SROC. Subgroup analyses were conducted for cancer type, sample type, and geographical region. Publication bias was assessed using Begg's and Egger's tests.Results: A total of 14 articles involving 18 studies and 1,526 participants (972 cases and 554 controls) were included. The diagnostic accuracy of miRNA-1246 in GICs was as follows: pooled sensitivity: 0.81 (95% CI: 0.79 - 0.83), specificity: 0.74 (95% CI: 0.71 - 0.77), PLR: 3.315 (95% CI: 2.33 - 4.72), NLR: 0.221 (95% CI: 0.153 - 0.319), DOR: 16.87 (95% CI: 9.45 - 30.09), AUC: 0.891, and Q*-index: 0.807. No publication bias was found based on Begg's (p = 0.172) and Egger's (p = 0.113) tests.Conclusion: Circulating miR-1246 shows promise as a non-invasive biomarker for early detection of GICs.

Risk of cancer and serious disease in Danish patients with urgent referral for serious non-specific symptoms and signs of cancer in Funen 2014-2021

BACKGROUND

In 2011, as the first European country, Denmark introduced the non-organ-specific cancer patient pathway (CPP) for patients presenting with non-specific symptoms and signs of cancer (NSSC). The proportion of patients with cancer over time is unknown.

METHODS

A retrospective cohort study of all patients with a NSSC-CPP investigational course in the province of Funen to the Diagnostic Centre in Svendborg from 2014 to 2021 was performed to evaluate the proportion of patients with cancer and serious disease over time.

RESULTS

A total of 6698 patients were referred to the NSSC-CPP of which 20.2% had cancer. While the crude referral rate increased from 114 per 100,000 people in 2014 and stabilised to around 214 in 2017-2021, the cancer detection rate of the total yearly new cancers in Funen diagnosed through the NSSC-CPP in DC Svendborg increased from 3 to 6%.

CONCLUSIONS

With now high and stable conversion and crude referral rates, the NSSC-CPP is one of the largest CPPs in Denmark as measured by the number of new cancer cases found. Similar urgent referral programmes in other countries might fill an unmet medical need for patients presenting with serious non-specific symptoms and signs of cancer in general practice.

Pre-Referral Primary Care Blood Tests and Symptom Presentation before Cancer Diagnosis: National Cancer Diagnosis Audit Data

BACKGROUND

Blood tests can support the diagnostic process in primary care. Understanding how symptomatic presentations are associated with blood test use in patients subsequently diagnosed with cancer can help to benchmark current practices and guide interventions.

METHODS

English National Cancer Diagnosis Audit data on 39,751 patients with incident cancer in 2018 were analysed. The frequency of four generic (full blood count, urea and electrolytes, liver function tests, and inflammatory markers) and five organ-specific (cancer biomarkers (PSA or CA125), serum protein electrophoresis, ferritin, bone profile, and amylase) blood tests was described for a total of 83 presenting symptoms. The adjusted analysis explored variation in blood test use by the symptom-positive predictive value (PPV) group.

RESULTS

There was a large variation in generic blood test use by presenting symptoms, being higher in patients subsequently diagnosed with cancer who presented with nonspecific symptoms (e.g., fatigue 81% or loss of appetite 79%), and lower in those who presented with alarm symptoms (e.g., breast lump 3% or skin lesion 1%). Serum protein electrophoresis (reflecting suspicion of multiple myeloma) was most frequently used in cancer patients who presented with back pain (18%), and amylase measurement (reflecting suspicion of pancreatic cancer) was used in those who presented with upper abdominal pain (14%). Prostate-specific antigen (PSA) use was greatest in men with cancer who presented with lower urinary tract symptoms (88%), and CA125 in women with cancer who presented with abdominal distention (53%). Symptoms with PPV values between 2.00-2.99% were associated with greater test use (64%) compared with 52% and 51% in symptoms with PPVs in the 0.01-0.99 or 1.00-1.99% range and compared with 42% and 31% in symptoms with PPVs in either the 3.00-4.99 or ≥5% range (p < 0.001).

CONCLUSIONS

Generic blood test use reflects the PPV of presenting symptoms, and the use of organ-specific tests is greater in patients with symptomatic presentations with known associations with certain cancer sites. There are opportunities for greater blood test use in patients presenting with symptoms that do not meet referral thresholds (i.e., <3% PPV for cancer) where information gain to support referral decisions is likely greatest. The findings benchmark blood test use in cancer patients, highlighting opportunities for increasing use.

Development and Internal Validation of a Risk Prediction Model to Identify Myeloma Based on Routine Blood Tests: A Case-Control Study

Myeloma is one of the hardest cancers to diagnose in primary care due to its rarity and non-specific symptoms. A rate-limiting step in diagnosing myeloma is the clinician considering myeloma and initiating appropriate investigations. We developed and internally validated a risk prediction model to identify those with a high risk of having undiagnosed myeloma based on results from routine blood tests taken for other reasons. A case-control study, based on 367 myeloma cases and 1488 age- and sex-matched controls, was used to develop a risk prediction model including results from 15 blood tests. The model had excellent discrimination (C-statistic 0.85 (95%CI 0.83, 0.89)) and good calibration (calibration slope 0.87 (95%CI 0.75, 0.90)). At a prevalence of 15 per 100,000 population and a probability threshold of 0.4, approximately 600 patients would need additional reflex testing to detect one case. We showed that it is possible to combine signals and abnormalities from several routine blood test parameters to identify individuals at high-risk of having undiagnosed myeloma who may benefit from additional reflex testing. Further work is needed to explore the full potential of such a strategy, including whether it is clinically useful and cost-effective and how to make it ethically acceptable.

Primary care blood tests before cancer diagnosis: National Cancer Diagnosis Audit data

BACKGROUND

Blood tests can support the diagnostic process in patients with cancer but how often they are used is unclear.

AIM

To explore use of common blood tests before cancer diagnosis in primary care.

DESIGN AND SETTING

English National Cancer Diagnosis Audit data on 39 752 patients with cancer diagnosed in 2018.

METHOD

Common blood test use (full blood count [FBC], urea and electrolytes [U&E], and liver function tests [LFTs]), variation by patient and symptom group, and associations with the primary care interval and the diagnostic interval were assessed.

RESULTS

At least one common blood test was used in 41% (n = 16 427/39 752) of patients subsequently diagnosed with cancer. Among tested patients, (n = 16 427), FBC was used in 95% (n = 15 540), U&E in 89% (n = 14 555), and LFTs in 76% (n = 12 414). Blood testing was less common in females (adjusted odds ratio versus males: 0.92, 95% confidence interval [CI] = 0.87 to 0.98) and Black and minority ethnic patients (0.89, 95% CI = 0.82 to 0.97 versus White), and more common in older patients (1.12, 95% CI = 1.06 to 1.18 for ≥70 years versus 50-69 years). Test use varied greatly by cancer site (melanoma 2% [ n = 55/2297]; leukaemia 84% [ n = 552/661]). Fewer patients presenting with alarm symptoms alone were tested (24% [ n = 3341/13 778]) than those with non-alarm symptoms alone (50% [ n = 8223/16 487]). Median primary care interval and diagnostic interval were longer in tested than non-tested patients (primary care interval: 10 versus 0 days; diagnostic interval: 49 versus 32 days, respectively, P<0.001 for both), including among tested patients with alarm symptoms (primary care interval: 4 versus 0 days; diagnostic interval: 41 versus 22 days).

CONCLUSION

Two-fifths of patients subsequently diagnosed with cancer have primary care blood tests as part of their diagnostic process. Given variable test use, research is needed on the clinical context in which blood tests are ordered.

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.

'Diagnostic downshift': clinical and system consequences of extrapolating secondary care testing tactics to primary care

Numerous drivers push specialist diagnostic approaches down to primary care ('diagnostic downshift'), intuitively welcomed by clinicians and patients. However, primary care's different population and processes result in under-recognised, unintended consequences. Testing performs poorer in primary care, with indication creep due to earlier, more undifferentiated presentation and reduced accuracy due to spectrum bias and the 'false-positive paradox'. In low-prevalence settings, tests without near-100% specificity have their useful yield eclipsed by greater incidental or false-positive findings. Ensuing cascades and multiplier effects can generate clinician workload, patient anxiety, further low-value tests, referrals, treatments and a potentially nocebic population 'disease' burden of unclear benefit. Increased diagnostics earlier in pathways can burden patients and stretch general practice (GP) workloads, inducing downstream service utilisation and unintended 'market failure' effects. Evidence is tenuous for reducing secondary care referrals, providing patient reassurance or meaningfully improving clinical outcomes. Subsequently, inflated investment in per capita testing, at a lower level in a healthcare system, may deliver diminishing or even negative economic returns. Test cost poorly represents 'value', neglecting under-recognised downstream consequences, which must be balanced against therapeutic yield. With lower positive predictive values, more tests are required per true diagnosis and cost-effectiveness is rarely robust. With fixed secondary care capacity, novel primary care testing is an added cost pressure, rarely reducing hospital activity. GP testing strategies require real-world evaluation, in primary care populations, of all downstream consequences. Test formularies should be scrutinised in view of the setting of care, with interventions to focus rational testing towards those with higher pretest probabilities, while improving interpretation and communication of results.

Identifying opportunities for timely diagnosis of bladder and renal cancer via abnormal blood tests: a longitudinal linked data study

BACKGROUND

Understanding pre-diagnostic test use could reveal diagnostic windows where more timely evaluation for cancer may be indicated.

AIM

To examine pre-diagnostic patterns of results of abnormal blood tests in patients with bladder and renal cancer.

DESIGN AND SETTING

A retrospective cohort study using primary care and cancer registry data on patients with bladder and renal cancer who were diagnosed between April 2012 and December 2015 in England.

METHOD

The rates of patients with a first abnormal result in the year before cancer diagnosis, for 'generic' (full blood count components, inflammatory markers, and calcium) and 'organ-specific' blood tests (creatinine and liver function test components) that may lead to subsequent detection of incidental cancers, were examined. Poisson regression was used to detect the month during which the cohort's rate of each abnormal test started to increase from baseline. The proportion of patients with a test found in the first half of the diagnostic window was examined, as these 'early' tests might represent opportunities where further evaluation could be initiated.

RESULTS

Data from 4533 patients with bladder and renal cancer were analysed. The monthly rate of patients with a first abnormal test increased towards the time of cancer diagnosis. Abnormalities of both generic (for example, high inflammatory markers) and organ-specific tests (for example, high creatinine) started to increase from 6-8 months pre-diagnosis, with 25%-40% of these patients having an abnormal test in the 'early half' of the diagnostic window.

CONCLUSION

Population-level signals of bladder and renal cancer can be observed in abnormalities in commonly performed primary care blood tests up to 8 months before diagnosis, indicating the potential for earlier diagnosis in some patients.

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.

Could Ovarian Cancer Prediction Models Improve the Triage of Symptomatic Women in Primary Care? A Modelling Study Using Routinely Collected Data

Simple Summary

Earlier detection of ovarian cancer has the potential to improve patient outcomes, including survival. However, determining which women presenting in primary care to refer for specialist assessment and investigation is a clinical dilemma. In this study, we used routinely collected English primary care data from 29,962 women with symptoms of possible ovarian cancer who were tested for the ovarian cancer biomarker CA125. We developed diagnostic prediction models to estimate the probability of the disease. A relatively simple model, consisting of age and CA125 level, performed well for the identification of ovarian cancer. Including additional risk factors within the model did not materially improve model performance. Following further validation, this model could be used to help triage symptomatic women in primary care based on their risk of undiagnosed ovarian cancer, identifying those at high risk for urgent specialist investigation and those at lower (but still elevated) risk for non-urgent investigation or monitoring.

Abstract

CA125 is widely used as an initial investigation in women presenting with symptoms of possible ovarian cancer. We sought to develop CA125-based diagnostic prediction models and to explore potential implications of implementing model-based thresholds for further investigation in primary care. This retrospective cohort study used routinely collected primary care and cancer registry data from symptomatic, CA125-tested women in England (2011–2014). A total of 29,962 women were included, of whom 279 were diagnosed with ovarian cancer. Logistic regression was used to develop two models to estimate ovarian cancer probability: Model 1 consisted of age and CA125 level; Model 2 incorporated further risk factors. Model discrimination (AUC) was evaluated using 10-fold cross-validation. The sensitivity and specificity of various model risk thresholds (≥1% to ≥3%) were compared with that of the current CA125 cut-off (≥35 U/mL). Model 1 exhibited excellent discrimination (AUC: 0.94) on cross-validation. The inclusion of additional variables (Model 2) did not improve performance. At a risk threshold of ≥1%, Model 1 exhibited greater sensitivity (86.4% vs. 78.5%) but lower specificity (89.1% vs. 94.5%) than CA125 (≥35 U/mL). Applying the ≥1% model threshold to the cohort in place of the current CA125 cut-off, 1 in every 74 additional women identified had ovarian cancer. Following external validation, Model 1 could be used as part of a ‘risk-based triage’ system in which women at high risk of undiagnosed ovarian cancer are selected for urgent specialist investigation, while women at ‘low risk but not no risk’ are offered non-urgent investigation or interval CA125 re-testing. Such an approach has the potential to expedite ovarian cancer diagnosis, but further research is needed to evaluate the clinical impact and health–economic implications.

Individual inflammatory marker abnormalities or inflammatory marker scores to identify primary care patients with unexpected weight loss for cancer investigation?

Background

Combinations of inflammatory markers are used as prognostic scores in cancer patients with cachexia. We investigated whether they could also be used to prioritise patients attending primary care with unexpected weight loss for cancer investigation.

Methods

We used English primary care electronic health records data linked to cancer registry data from 12,024 patients with coded unexpected weight loss. For each individual inflammatory marker and score we estimated the sensitivity, specificity, likelihood ratios, positive predictive value (PPV) and the area under the curve along with 95% confidence intervals for a cancer diagnosis within six months.

Results

The risk of cancer associated with two abnormal inflammatory markers combined in a score was higher than the risk associated with individual inflammatory marker abnormalities. However, the risk of cancer in weight loss associated with individual abnormalities, notably a raised C-reactive protein, was sufficient to trigger further investigation for cancer under current NICE guidelines.

Conclusions

If scores including pairs of inflammatory marker abnormalities were to be used, in preference to individual abnormalities, fewer people would be investigated to diagnose one cancer with fewer false positives, but fewer people with cancer would be diagnosed overall.

MicroRNA-223 in gastrointestinal cancers: A systematic review and diagnostic meta-analysis

BACKGROUND

Several studies have been conducted on the diagnostic role of miR-223 in cancers related to the digestive system. However, the diagnostic role of this microRNA in gastrointestinal (GI) cancers has not been fully elucidated. This meta-analysis aimed to accurately assess the diagnostic role of circulating miR-223 in GI cancers.

METHODS

A literature search was performed in PubMed/Medline, Science Direct, Web of Science, Google Scholar, Embase and Scopus, up to 1st May 2020 databases. Twelve studies were eligible and included in the analysis. Meta-Disc software was used to calculate the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, area under the curve (AUC) and the summary receiver operating characteristic (SROC) based on true positive, true negative, false negative and false positive for each gastrointestinal cancer separately and in total.

RESULTS

Twelve case-control studies were included with 1859 participants (1080 cases and 779 controls). Pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio were 0.77 (95% CI: 0.74-0.79), 0.75 (95% CI: 0.72-0.78), 3.04 (95% CI: 2.20-4.18), 0.31 (95% CI: 0.22-0.42) and 10.77 (95% CI: 5.96-19.47), respectively. AUC was 0.83, suggesting a high-grade diagnostic precision of miR-223 in gastrointestinal cancers. Besides, subgroup analyses were performed to assess the diagnostic power of miR-223 based on the type of gastrointestinal cancer, sample type and country via calculating pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio.

CONCLUSION

Our meta-analysis showed the value of circulating miR-223 levels in the early diagnosis of diverse digestive system carcinomas.