Real-world data evaluating Guy’s rapid diagnostic clinic as an alternate pathway for patients with FIT levels below 10

Abstract

Objective

To analyse the effectiveness of rapid diagnostic clinics (RDCs) as an alternative pathway for patients with concerning symptoms and a faecal immunochemical test (FIT) result <10. Our primary endpoint was rate of colorectal cancer (CRC) detection. Second endpoints were rates of other cancers and gastrointestinal (GI) serious benign conditions. Finally, we analysed the specific pathway followed by FIT <10 patients with cancer at Guy’s and St Thomas NHS Foundation Trust (GSTT) RDC.

Design

A retrospective and prospective cohort study.

Setting

GSTT RDC, one of England’s largest single-centre RDCs. Sociodemographic and clinical characteristics of FIT <10 patients were analysed descriptively.

Participants

Patients with an FIT result <10, seen at GSTT RDC between 1 January 2020 and 5 May 2023.

Results

A total of 1299 patients with an FIT<10 were seen at GSTT RDC between January 2020 and May 2023. Of these, 66% (n=861) reported weight loss, 62% (n=805) pain, 37% (n=481) fatigue, 34% (n=444) were anaemic and 23% (n=301) had nausea and vomiting. Among these patients, 7% (n=88) received a cancer diagnosis, 36% (n=462) were identified as having a serious benign condition. Within the patients with cancer, 9% (n=8) were diagnosed with CRC. Among patients with serious benign conditions, 7% (n=31) were referred to colorectal, hepatopancreatobiliary, or upper GI specialists.

Conclusion

This study demonstrates the effectiveness of RDCs as an alternate pathway for FIT <10 patients with ongoing clinical concerns. These results contribute to enhancing patient care and optimising resource allocation within the healthcare system.

Strengths and limitations of this study.

• To our knowledge, this is the first descriptive study investigating the effectiveness of rapid diagnostic clinics (RDCs) as an alternative pathway for patients presenting concerning symptoms and negative FIT<10.

• The Guy’s and St. Thomas NHS Foundation Trust (GSTT) RDC is one of the UK’s largest single centre RDCs, offering substantial real-world data for analysis.

• Our study uses real-world data to analyse true patient experiences, enabling healthcare providers to make informed decisions that better align with patient needs.

• Our study includes data exclusively from GSTT RDC, limiting the generalisability of findings to other RDC settings. However, the diverse patient population at GSTT enhances the representativeness of our findings within this context.

• Our study lacks long-term follow-up data, preventing our ability to ascertain if patients developed cancer at a later stage.

Introduction

Evaluating symptomatic patients with a potential colorectal cancer is often challenging within primary care. These patients often have symptoms, such as abdominal discomfort and iron deficiency anaemia (IDA), that may be indicative of gynaecological, upper gastrointestinal (UGI) or haematological conditions, making the diagnosis complex.¹ The faecal immunochemical test (FIT) was introduced as a non-invasive and cost-effective screening tool for detecting possible colorectal cancer (CRC) in primary care. It is a quantitative stool test that detects an intact globin component of haemoglobin and can reliably measure faecal haemoglobin (fHb) concentration.² Unlike with other faecal occult blood tests, FIT offers improved sensitivity and specificity which has led to its widespread adoption in CRC screening programmes.³

FIT was first recommended by the National Institute for Health and Care Excellence (NICE) in 2017 to simplify CRC screening in the United Kingdom (UK). These original NICE guidelines (DG30) recommended the FIT to adults with low-risk symptoms (online supplemental table 1), and a positive test (>10 ug) was considered a qualifying criterion for the 2-week wait (2WW) CRC pathway.⁴ The implementation of FIT gained further traction during the COVID-19 pandemic, where it enabled the prioritisation of patients who met the 2WW criteria for colonoscopy, while alleviating pressure on endoscopy services.⁵ Furthermore, in 2021 the NICE FIT study, which aimed to assess whether FIT can be used to select patients with 2WW symptoms for urgent investigations, suggested that the diagnostic accuracy of FIT for CRC is superior to high-risk 2WW symptoms outlined in online supplemental table 1. Therefore, FIT can be used to prioritise patients for investigations, as CRC is more likely to occur in those with higher fHb concentrations.⁶ Consequently, the NICE NG12 CRC guidelines were updated, establishing FIT as the standard of care for CRC referrals into urgent 2WW CRC pathways.⁷ Similarly, a joint guideline from the Association of Coloproctology of Great Britain and Ireland (ACPGBI) and the British Society of Gastroenterology (BSG) was published. In summary, these guidelines recommend that: (1) FIT positive (>10 ug) patients should be referred onto a 2WW pathway, (2) patients with an FIT <10 ug with no ongoing clinical concern should be managed within primary care and (3) FIT <10 ug patients but with ongoing clinical concern should still be referred onto either a routine or urgent pathway, such as a non-site-specific (NSS) pathway (online supplemental figure 1).⁸

Rapid diagnostic clinics (RDCs) have been established to expedite cancer diagnoses and improve patient’s experience in patients with vague symptoms concerning of cancer (eg, weight loss, fatigue, vague abdominal pain), who do not align to specific cancer pathways or fulfil multiple 2WW pathway referral criteria.⁹ RDCs provide a streamlined pathway by prioritising diagnostics while prehabilitating patients by managing comorbidities, polypharmacy, nutrition and mental health conditions. There are various typologies of RDCs depending on the clinicians involved (eg, gastroenterology, medicine, oncology and general practitioners (GP)) and the model (straight to test radiology vs clinical models vs primary care models). The typology is usually decided by the local staff clinical skill set, ability to recruit and geographic area they cover. The Guy’s and St. Thomas NHS Foundation Trust (GSTT) RDC was established in 2016 and has seen over 5000 patients to date. We run a holistic multidisciplinary model, which includes a blend of primary and secondary care specialists, such as consultants in oncology, acute medicine and general practice.¹⁰ Diagnostics are performed within 1–2 weeks to facilitate expedited diagnosis within 1–3 weeks, with prompt referral or discharge. Our local evaluation reported a cancer detection rate of 7.2% at GSTT RDC, which is higher than many 2WW pathways, including a rate of 3% within our local CRC pathway.⁹

This study aimed to analyse the effectiveness of RDCs as an alternative pathway for patients with concerning symptoms and an FIT <10. Our primary endpoint was rate of CRC detection. Second endpoints were to evaluate the rates of other cancers and gastrointestinal (GI) serious benign conditions. Finally, we analysed the specific pathway followed by FIT <10 patients with cancer at GSTT RDC, showcasing the successful redirection of patients to appropriate NSS pathways.

Materials and methods

Rapid diagnostic clinic at GSTT

The RDC at GSTT offers an efficient pathway for adult patients (over 18) in southeast London presenting with vague symptoms suggestive of cancer or serious benign conditions, who do not qualify for other site-specific urgent pathways or fall into multiple diagnostic categories. During the initial consultation, patients undergo a comprehensive assessment encompassing symptom evaluation, physical comorbidities, polypharmacy, lifestyle, social factors, mental health and nutrition. As previously mentioned, this assessment is conducted through a holistic, multidisciplinary approach involving consultants in oncology, acute medicine and general practice. Diagnostic tests, including blood tests, imaging and endoscopies, are performed within 2 weeks of the consultation. The overarching aim of the RDC is to expedite diagnosis, facilitate further evaluation as needed or discharge patients back to their GP within a 28-day timeframe.⁹

Study population

Our study population were patients referred to GSTT RDC between 1 January 2020 and 5 May 2023. Data were collected retrospectively from the electronic patient records (EPR) for patients seen between 1 January 2020 and 31 May 2021. For patients seen from 1 June 2021 until 5 May 2023, data were collected prospectively and managed using REDCap electronic data capture tools hosted at King’s College London.¹¹ The study cohort had an average follow-up duration of 1.5 years. Data on FIT results, gender, ethnicity, socioeconomic status (SES), age, performance status, symptoms (ie, number, type and duration), smoking status, comorbidities, tests undertaken and outcomes were collected. Based on previously mentioned guidelines, we defined a negative FIT as ≤10 µg Hb/g faeces.⁸ SES was based on the index of multiple deprivation (IMD) decile score,¹² where low SES was defined as a score of ≤3, middle as 4–7, and high as ≥8. Cancer diagnoses were confirmed by histology or radiology. Serious benign conditions were defined as those requiring onward secondary care referral. Patients were excluded from analysis if opted out of their confidential patient information being used for research using the NHS Digital National data opt out programme. All data were collected and analysed as part of GSTT Cancer Cohort.

Statistical analysis

Data on sociodemographic and clinical characteristics for patients with a negative FIT were analysed in a descriptive manner. Data management and analysis were carried out using StataIC V.15.1 (College Station, Texas). Data are presented in line with the Strengthening the Reporting of Observational Studies statement for cohort studies (online supplemental table 2).¹³

Patient and public involvement

No patient is involved.

Results

A total of 3319 patients were seen at GSTT RDC between January 2020 and May 2023. Out of these, 41% (n=1376) had an FIT at referral, of which 1299 (94%) had a negative result (figure 1). Table 1 presents the patient characteristics of those who had an FIT <10 test: 41% (n=537) of patients were men. The median age was 63 years, with an IQR of 52–74 years. Forty-six per cent (n=596) of patients identified as white, 20% (n=253) as black, 6% (n=83) as Asian, and 5% (n=60) as mixed or other ethnicities. Thirty-eight per cent of patients (n=489) were classified as having a low SES, 45% (n=590) as middle and 17% (n=215) as high. As for clinical characteristics, 75% (n=974) of patients exhibited a PS of 0–1, 16% (n=211) had a PS of 2, and 5% (n=67) had a PS of 3 or higher. The median number of symptoms experienced by patients in this cohort was 2 (IQR of 2 to 3).

Figure 1.

Study population. FIT, faecal immunohistochemical test.

Table 1.

FIT negative patient characteristics

	FIT<10 ug n=1299 n (%)
Gender
Male	537 (41)
Ethnicity
White	596 (46)
Black	253 (20)
Asian	83 (6)
Mixed and Other	60 (5)
Not known	307 (23)
Socioeconomic status (IMD)
Low	489 (38)
Middle	590 (45)
High	215 (17)
Age
Median (IQR)	63 (52–74)
Performance status
0–1	974 (75)
2	211 (16)
3+	67 (5)
Number of symptoms
Median (IQR)	2 (2-3)
Symptoms
Weight loss	861 (66)
Fatigue	481 (37)
Nausea and vomiting	301 (23)
Progressive pain	805 (62)
Anaemia	444 (34)
Symptom duration
<6 months	652 (50)
≥6 months	526 (41)
Smoking status
Current	249 (19)
Former	298 (23)
Never	249 (19)
Obesity
Yes	192 (15)

FIT, faecal immunohistochemical test; IMD, index of multiple deprivation; IQR, Interquartile range.

During the analysis of symptoms reported at referral, several key findings emerged. A considerable proportion of patients 66% (n=861) reported weight loss, 62% (n=805) pain, 37% (n=481) fatigue, 34% (n=444) were anaemic and 23% (n=301) had nausea and vomiting. Fifty per cent of referred patients had experienced symptoms for less than 6 months. Nineteen per cent (n=249) were current, 23% (n=298) former and 19% (n=249) never smokers.

Table 2 presents the primary diagnoses of FIT-negative patients in detail. Among these patients, 7% (n=88) received a cancer diagnosis, while 36% (n=462) were identified as having a serious benign condition requiring onward referral to a hospital specialist. The remaining 57% (n=750) were referred back to their general practitioner (GP) with a mild benign condition.

Table 2.

Primary diagnoses of patients with FIT negative tests

Primary diagnosis	FIT<10 ug n=1299 n (%)
Cancer	88 (7)
Cancer grouping
Colorectal	8 (9)
Hepatobiliary	14 (16)
Haematology	13 (15)
Lung	13 (15)
Urology	13 (15)
Breast	8 (9)
Head and neck	7 (8)
Neuroendocrine tumour	5 (6)
Upper gastrointestinal	2 (2)
Sarcoma	2 (2)
Melanoma	1 (1)
Meningioma	1 (1)
Adrenal	1 (1)
Serious benign condition	462 (36)
Specialty of referral
Colorectal	4 (1)
Hepatobiliary	16 (4)
Upper gastrointestinal	11 (2)
Mild benign condition	750 (57)

FIT, faecal immunohistochemical test.

Within the cancer diagnosis group (n=88), 9% (n=8) were diagnosed with colorectal cancer (representing 0.6% of the overall FIT-negative population), with 50% (n=4) being a right-sided CRC. Sixteen per cent (n=14) of patients within the cancer group had a hepato-pancreato-biliary (HPB) cancer, 15% (n=13) haematology (including a case of gastric diffuse large B-cell lymphoma). Lung cancer, urological cancers and breast cancer were each observed in 15% (n=13) of cases. Head and neck malignancies were present in 8% (n=7) of cases, while neuroendocrine tumours were seen in 6% (n=5) of cases. UGI and sarcoma cancers accounted for 2% (n=2) each. Additionally, 1% (n=1) of cases were diagnosed with melanoma, 1% (n=1) with meningioma, and 1% (n=1) with adrenal malignancy.

Moreover, among patients with serious benign conditions, only 1% (n=4) were referred to colorectal specialists, 4% (n=16) were referred to HPB specialists and 2% (n=11) were referred to UGI specialists (including other GI-related specialties). The remaining patients were redirected to non-GI pathways.

Finally, table 3 illustrates that out of the total 31 patients who were referred to a secondary care GI specialist (including colorectal, HPB, UGI and UGI-related specialties), 51.6% (n=16) received a diagnosis related to HPB conditions, 19.4% (n=6) were diagnosed with UGI conditions, 12.9% (n=4) had a diagnosis of colorectal condition (specifically, inflammatory bowel disease) and 16.1% (n=5) were diagnosed by other GI-related specialists (eg, interstitial lung disease and polymyalgia rheumatica).

Table 3.

Final benign diagnoses for patients with negative FIT referred to colorectal, hepatobiliary, upper GI or other gastrointestinal-related specialties

Colorectal (n=4) 12.9%
Inflammatory bowel disease	4 (100%)
Hepatobiliary (n=17) 51.6%
Chronic liver disease	9 (37.5%)
Hepatitis B	2 (12.5%)
Pancreatic insufficiency	1 (6.3%)
Portal hypertension	1 (6.3%)
Liver/pancreato-biliary lesions	6 (37.5%)
Upper GI (n=6) 19.4%
Barret’s oesophagus	3 (50%)
Peptic ulcer disease	2 (33.3%)
Gastric volvulus	1 (6.7%)
Other (n=5) 16.1%
Interstitial lung disease	1 (20%)
Polymyalgia rheumatica	1 (20%)
Adrenal adenoma	1 (20%)
Epiglottic lesion	1 (20%)
Cricopharyngeal spasm	1 (20%)

FIT, faecal immunohistochemical test; GI, gastrointestinal.

Finally, figure 2 provides a comprehensive overview of the pathway followed by patients who received a cancer diagnosis within our cohort. The vast majority (96%) were initially referred from primary care, while the remaining 4% consisted of internal referrals, including one patient referred from colorectal services. The three most prevalent symptoms reported were weight loss (60% (n=53)), anaemia (54% (n=47)) and abdominal pain (23% (n=20)). Among the 88 patients diagnosed with cancer, only 20% (n=17) underwent an endoscopic procedure, while 95% (n=92) of patients underwent at least one imaging study. Specifically, 85% of patients had a CT with contrast of the abdomen and pelvis (CT Cap) and 20% underwent an abdominal ultrasound.

Figure 2.

Pathway of FIT-negative patients leading to subsequent cancer diagnosis. CIBH, change in bowel habit; CR, colorectal; CT CAP, CT chest abdomen pelvis; FIT, faecal immunohistochemical test; GP, general practitioner; US, ultrasound.

Discussion

A total of 1299 FIT <10 patients were successfully redirected from the 2WW CRC to the GSTT RDC over 3 years. Out of this cohort, only 55 patients were rereferred to a GI service, comprising of 24 cases of confirmed cancer and 31 possible benign conditions. Additionally, the observed rate of CRC among FIT <10 patients was 0.6%, aligning closely with the 0.4% reported in the NICE FIT study.⁶ These findings, therefore, support the NHS guidance, based on the BSG/ACPGBI FIT recommendations, to use FIT as a triage test to redirect patients with an FIT <10 but ongoing clinical concern towards a NSS pathway.

A systematic review and meta-analysis of 49 studies looking at key outcomes of the UK’s 2WW CRC pathway, before the introduction of FIT testing, reported an unexpectedly low CRC conversion rate (pooled rate of 7.7%).¹⁴ This affirmed that stratification of patient referrals based on symptoms alone, regardless of how quickly they were seen, did not improve CRC detection. Moreover, since the introduction of the revised NG12 2WW guidelines for cancer in 2015,¹⁵ there has been an exponential increase in CRC referrals. A recent study by Faux et al looking at data from the Royal Cornwall Hospital reported that while referrals in other 2WW services increased by 2%, the CRC 2WW service grew by 90% from 2015 to 2019. The same study also reported that since its introduction, a negative FIT in symptomatic patients signified a low risk of CRC; however, overall cancer prevalence for this group remained over 6%.¹⁶ These findings are in line with those from our current study where non-colorectal malignancies accounted for 6.1% of our FIT<10 cohort.

Furthermore, since its publication, the BSG/ACPGBI recommendations have helped to risk-stratify patients based on their FIT results, and many patients could now be considered eligible for referral to a NSS service.⁸ RDCs, among other NSS services, allow fluidity between site and non-site-specific pathways for patients with symptoms concerning of malignancy and streamline patient care while also alleviating the ongoing pressure on 2WW CRC services and consequently, endoscopic services.^{9 17} In this context, when comparing the endoscopy rates of GSTT RDC with the 2WW CRC pathway, we found that RDC rates are significantly lower (25% vs ~75%, respectively (data not presented)). On the other hand, a large proportion (85%) of patients evaluated at GSTT RDC underwent a CT CAP, a more cost-effective and minimally invasive imaging modality, which enables a comprehensive assessment of other organs, which may lead to non-CRC diagnoses. Nevertheless, a major limitation of our FIT <10 pathway is that because only a small percentage of patients undergo endoscopic evaluation, so there is a risk that we may be missing premalignant conditions, such as polyps that could be removed or kept under surveillance prevent future development of CRC.¹⁸ With regard to this, various studies have reported that the sensitivity of FIT for serious benign bowel conditions, including advanced adenomas and IBD, is considerably lower than for CRC.⁸ In the case of advanced adenomas, a systematic review and meta-analysis of 31 studies published in 2019 reported a low sensitivity with a positive FIT (test threshold of 0.41 and CI of 0.33–0.47).¹⁹ In the case of IBD, NICE guidelines recommend faecal calprotectin (FCP) testing for the diagnosis of IBD.²⁰ A recent cohort study compared FCP to FIT and found that both studies have similar sensitivity, therefore, in case of clinical concern of IBD, FCP should always be considered first.²¹

Furthermore, various studies have investigated the benefits of repeated FIT testing.²² For instance, a recent cohort study looking at 5761 patients concluded that a second negative test within 12 months of the first sample, in the context of ongoing clinical concern reliably rules out CRC.²³ However, BSG/ACPGBI recommendations concluded that there is currently insufficient evidence to recommend the use of a repeat FIT to guide referrals in routine practice.⁸ It is worth noting that upcoming guidelines are likely to incorporate a second FIT, enhancing the ability to confidently exclude colorectal cancer.

Finally, findings from previous studies have suggested that FIT sensitivity is higher for left-sided CRCs and that right-sided lesions are more frequently missed by FIT.^{8 24 25} For instance, a recent UK study by Cunin et al looking at symptomatic patients undergoing FIT on a CRC pathway reported that of seven FIT <10 patients, six were right-sided (caecal) tumours, suggesting that the nature and biology of these tumours may play a role in their FIT sensitivity.²⁴ One explanation could be that the larger degradation of haemoglobin, which may occur at a greater extent in right-sided lesions, leads to lower concentrations of haemoglobin in the stool.²⁶ A second hypothesis could be that due to the distribution of blood throughout the formed stool also reduces the concentration of haemoglobin.⁸ Finally, most of the advanced serrated lesions are right sided and tend to bleed less than advanced adenomas.²⁵ This may partly explain the missed CRC rate within our cohort since 50% (n=4) of missed CRC were right-sided tumours.

Additionally, the observation that most patients with weight loss underwent CT CAP raises questions about the utility of providing direct access to CT CAP for GPs. Direct access to CT CAP for patients presenting with weight loss could potentially streamline the diagnostic process, facilitating timely evaluation and management decisions. This approach warrants further exploration and consideration in the context of optimising patient care pathways and resource allocation within the healthcare system.

To our knowledge, this is one of the first studies looking at the effectiveness of NSS services as an alternative pathway for FIT<10 patients with ongoing clinical concerns. As previously mentioned, findings from our study align with current NHS guidelines, affirming the use of FIT <10 as a triage tool into NSS pathways. By effectively managing these patients within the RDC setting, we contribute to the overall mitigation of pressure on urgent CRC pathways. Additionally, our study revealed that only a small proportion of patients required a rereferral to a GI team, indicating the successful management of FIT <10 patients within the RDC. The rate of missed CRC within our cohort was found to be 0.6%, aligning with current guidelines. However, it is imperative to establish a mechanism to identify and rerefer patients with ongoing clinical concern or CRC despite initial work-up in our RDC.

While our FIT <10 pathway demonstrates numerous advantages, it is important to acknowledge certain limitations. The reduced GI input and fewer endoscopies performed within the RDC setting may have potential long-term implications for CRC rates, as we might miss the detection of premalignant conditions that would have otherwise been identified through endoscopic procedures, subsequently leading to appropriate surveillance by the colorectal team. Another limitation arises from RDCs functioning within the 2 week wait model. When we identify patients who could benefit from non-urgent endoscopic procedures or enhanced GI input, the current protocol mandates referral back to their GP for redirection. This process may introduce delays and obstacles, potentially affecting the timeliness and efficiency of care delivery. For instance, an example would be those patients who present with anaemia without another concerning indicator. Currently, only 35% of patients with anaemia undergo endoscopic procedures, highlighting the potential gap in capturing cases that could benefit from further evaluation.

Conclusion

In conclusion, this study provides compelling evidence supporting the effectiveness of RDCs as an alternate pathway for FIT-negative patients with ongoing clinical concerns. Our findings validate the use of FIT as a triage tool and demonstrate the potential of RDCs in reducing the burden on urgent colorectal cancer pathways. However, further research and the establishment of appropriate mechanisms are warranted to address ongoing clinical concerns and potential missed opportunities for detecting premalignant conditions. These findings contribute to the continuous enhancement of patient care and inform strategies to optimise resource allocation within the healthcare system.

Data availability statement

Data are available upon reasonable request. The data presented in this study are available on request from the corresponding author. The data are not publicly available due to ethical reasons.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

All data were collected and analysed as part of GSTT Cancer Cohort (Ethics reference number: 23/NW/0105).