Advice
Clinical and technical evidence
Clinical and technical evidence
A literature search was carried out for this briefing in accordance with the interim process and methods statement for medtech innovation briefings. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.
Published evidence
Six studies are summarised in this briefing, including a total of 622 adults with inflammatory bowel disease (IBD) and 45 laboratory specimens.
The summarised evidence includes 4 prospective validation studies and 2 retrospective observational studies.
There is additional evidence not summarised in this briefing, including 2 in‑vitro studies (Ong et al. 2021a, Ong et al. 2021b, Ong et al. 2020, Yadav et al. 2021) and a small sample prospective validation study (Šahinović et al. 2021). Another study (Lorenzon et al. 2021) was identified in the literature search. But it was excluded as it included people with active inflammatory disease or infection, of whom only 17% had suspected or diagnosed IBD.
The clinical evidence and its strengths and limitations is summarised in the overall assessment of the evidence.
Overall assessment of the evidence
The evidence on ProciseDx is mostly validation studies on the 4 Procise assays. The evidence was reported in non-peer-reviewed preprints, abstracts and posters. It is limited in quality, with some studies lacking information on demographics, study design and findings. All studies compared the Procise assays with appropriate comparators. Findings showed that the Procise assays had comparable accuracy to established tests for IBD and good performance in detecting loss of response in people with IBD on maintenance therapy. ProciseDx was also found to be quicker, with results returned in around 3 minutes. But there was no evidence on the prospective use of ProciseDx in clinical care or therapeutic drug monitoring. Most studies included people with IBD and there was no evidence on using ProciseDx to diagnose the condition. None of the studies were done in the UK so there is no evidence on the use of the platform in the NHS. Further research should evaluate the use of ProciseDx in clinical care for diagnosing and monitoring IBD, including its effect on clinical decision making and dosing, resource use and patient outcomes.
Cerna et al. (2022)
Intervention and comparator
Procise IFX and Procise ADL using capillary finger prick compared with enzyme-linked immunosorbent assay (ELISA) test using peripheral venous samples.
Key outcomes
Spearman's correlation with ELISA assays was 0.94 (R squared 0.88, slope 0.93, intercept 0.30) for Procise IFX and 0.91 (R squared 0.83, slope 1.40, intercept -1.26) for Procise ADL (both p<0.001). Authors concluded that Procise IFX and Procise ADL are accurate and comparable to ELISA.
Strengths and limitations
The study was reported in an abstract and poster only. There was limited detail on study design and methods. It was unclear if the study included people with IBD and if all samples were taken prospectively. The abstract did not report if any samples were excluded from analysis.
Stevens et al. (2022a)
Study size, design and location
Retrospective observational study examining the clinical utility of Procise ADL in 84 adults with Crohn's disease or ulcerative colitis who had maintenance adalimumab treatment. The study compared adalimumab levels in people with (n=37) and without (n=47) loss of response. Study location was not reported.
Key outcomes
Procise ADL assay performance in detecting loss of response was optimised at adalimumab trough cut-off of 8 micrograms per millilitre (sensitivity 57%, specificity 89%). Area under the receiver operating characteristic (ROC) curve for loss of response was 0.82 (95% confidence interval [CI] 0.73 to 0.92). Adalimumab trough levels were lower in people who had loss of response (median 6.0 micrograms per millilitre) than those who did not (median 13.0 micrograms per millilitre; p<0.001). People with adalimumab levels below 8 micrograms per millilitre had a 5.34‑fold increased risk of loss of response than those with higher adalimumab levels.
Strengths and limitations
The study was reported in an abstract and poster only. Findings provide some support for therapeutic drug monitoring using Procise ADL to detect loss of response in IBD. But the retrospective study design meant the technology was not used in clinical practice. More research is therefore needed on the use of Procise ADL in clinical practice and its impact on clinical decision making and outcomes. The company was involved in the research.
Stevens et al. (2022b)
Study size, design and location
Retrospective observational study examining the clinical utility of Procise IFX in 92 adults with Crohn's disease or ulcerative colitis who had maintenance infliximab treatment. The study compared infliximab levels in people with (n=55) and without (n=37) loss of response. Study location was not reported.
Key outcomes
Procise IFX assay performance in detecting loss of response was optimised at infliximab trough cut-off of 3 micrograms per millilitre (sensitivity 64%, specificity 89%). Area under the ROC curve for loss of response was 0.82 (95% CI 0.73 to 0.91). Infliximab trough levels were lower in people who had loss of response (median 2.4 micrograms per millilitre) than those who did not (median 6.5 micrograms per millilitre; p<0.001). People with infliximab levels below 3 micrograms per millilitre had a 5.89‑fold increased risk of loss of response than those with higher infliximab levels.
Strengths and limitations
The study was reported in an abstract and poster only. Findings provide some support for therapeutic drug monitoring using Procise IFX to detect loss of response in IBD. But more research is needed on the use of Procise IFX in clinical practice. The company was involved in the research.
Maniero et al. (2021)
Study size, design and location
Prospective analytical validation study in 87 adults with IBD taking infliximab in Italy. Of these, 52% had Crohn's disease, 45% had ulcerative colitis and 3% had underdetermined IBD.
Intervention and comparator
Procise IFX assay using finger-prick whole blood compared with Promonitor infliximab ELISA test (Grifols) using serum samples.
Key outcomes
Procise IFX took about 3 minutes from blood collection to results. Serum sampling took about 3 weeks, and the ELISA test took 3 hours to do. Deming regression showed a correlation between the tests of 0.83 (R squared 0.69), with a slope of 1.44 and intercept of -1.36. Authors concluded Procise IFX had similar accuracy to ELISA and was quicker, and easy to use.
Strengths and limitations
The study was reported in an abstract only. Finger-prick whole blood and serum samples were collected at the same time, which may have reduced timing errors. The tests differ in the reportable range of infliximab levels. The reportable range for Procise IFX was 1.7 to 77.2 micrograms per millilitre, while ELISA ranged 0.04 to 14.4 micrograms per millilitre. In total, 39 people were excluded from analysis because they had trough levels outside of the range for either test. The company was involved in the research.
Marsilio et al. (2021)
Study size, design and location
Prospective analytical validation study in 60 adults with IBD taking adalimumab in Italy. Of these, 80% had Crohn's disease, 17% had ulcerative colitis and 3% had underdetermined IBD.
Intervention and comparator
Procise ADL assay using finger-prick whole blood compared with Promonitor adalimumab ELISA test (Grifols) using serum samples.
Key outcomes
Procise ADL took about 3 minutes for the results to be returned. Serum sampling took about 3 weeks, and the ELISA test took 3 hours to do. Deming regression showed a correlation between the tests of 0.86 (R squared 0.74), with a slope of 1.30 and intercept of -1.52. Authors concluded Procise ADL had similar accuracy to ELISA and was quicker, and easy to use.
Strengths and limitations
The study was reported in an abstract only. Finger-prick whole blood and serum samples were collected at the same time. The tests differ in the reportable range of adalimumab levels. The reportable range for Procise ADL was 1.3 to 51.5 micrograms per millilitre, while ELISA ranged from 0.02 to 12 micrograms per millilitre. In total, 30 people were excluded from comparative analysis because they had trough levels of adalimumab outside of the range for either test. The company was involved in the research.
Volkers et al. (2021a)
Study size, design and location
Prospective validation study in adults with IBD who needed routine measurement of C-reactive protein, infliximab and adalimumab. Study location was not reported. This study was also reported in a conference poster (Volkers et al. 2021b).
The study recruited 66 people for C‑reactive protein measurements, 124 for infliximab, and 109 for adalimumab. Some measurements were excluded from analysis because they were outside of the assay range. Data was analysed for 41 people having C‑reactive protein measurement, 120 having infliximab, and 105 having adalimumab.
Intervention and comparators
Procise CRP, Procise IFX and Procise ADL assays using finger-prick whole blood samples compared with laboratory tests using C-reactive protein plasma assay, and infliximab and adalimumab serum ELISA. Procise IFX was also done using serum samples because of outliers in the findings.
Key outcomes
ProciseDx results were returned in 3 minutes, but the time for comparator results was not reported. Procise CRP, Procise IFX and Procise ADL using finger-prick whole blood samples correlated with laboratory tests (0.98, 0.88, 0.87, respectively; all p<0.001). Deming regression analysis of C-reactive protein assays resulted in a slope of 0.71 (95% CI 0.5 to 0.9) and intercept of 1.5 (95% CI -0.4 to 3.5). The authors reported there was an underestimation of C‑reactive protein but considered this clinically irrelevant. Infliximab assays had a slope of 1.1 (95% CI 0.83 to 1.3) and intercept of 1.4 (95% CI -0.5 to 3.4), while adalimumab assays had a slope of 1 (95% CI 0.9 to 1.2) and intercept of 1.9 (95% CI 0.5 to 3.2). Procise IFX using serum samples correlated with ELISA (0.99, p<0.001), with Deming regression analysis resulting in a slope of 1.1 (95% CI 1.0 to 1.1) and intercept of 0.95 (95% CI 0.4 to 1.5). Authors suggested outliers between infliximab finger-prick whole blood and serum samples may be because of timing errors.
Strengths and limitations
The study was reported in a poster and abstract only. There was limited detail, including little information on study design including sampling or blinding, and study location. There was also limited discussion of findings, particularly considerations related to the underestimation of C‑reactive protein and timing errors. The study was funded by the company and involved 2 employees.
Sustainability
The company claims that the ProciseDx point-of-care platform will reduce the need for biological samples to be packaged and transported from clinical sites to laboratories for processing. There is currently no evidence on this.
Recent and ongoing studies
Clinical validation of a point of care (POC) test for the measurement of infliximab (IFX) or adalimumab (ADL) levels in the serum of inflammatory bowel disease (IBD) patients. Trialregister.nl identifier: NL8934. Status: recruiting. Indication: inflammatory bowel disease. Devices: Procise IFX and Procise ADL assays. Date registered: September 2020. Country: Netherlands.