Proficiency Testing as a tool to highlight issues with sample processing when analyzing stool samples to investigate infectious diarrhea using multiplex assays.

In January 2023 the British Columbia Ministry of Health under the Laboratory Services Act approved a pathogen multiplex panel fee code item to encourage diagnostic microbiology laboratories to move towards syndromic testing.
In response to this change in the diagnostic approach, CMPT developed a PT panel consisting of simulated/clinical samples targeting the 14 organisms listed in the BC Infectious Diarrhea – Guideline for Investigation that came into effect October 27, 2022.[1]

The new gastrointestinal panel (GIP) PT scheme was designed to meet DAP requirements for EQA for these regulated organisms and samples shipped twice per year to subscribing laboratories. Subscription to CMPT’s PT scheme was at the discretion of the subscriber’s choice, and CMPT enrolled 13 laboratories in the first year. Of the 13 subscribing laboratories, 7 used Biofire Film Array (bioMérieux), 4 used Allplex (Seegene), 1 used a BD Max system (BD) and one lab did not disclose their platform.

Methods

The source of material for PT samples was donated patients’ stools in Cary Blair stored at -20C.  In order to prepare enough volume to send to participants the original samples were diluted 10X with PBS and then an extra 2x dilution in stool matrix (final dilution: 20X). The PT samples were sent for confirmation to a reference laboratory before sending them to participant laboratories.

These samples were characterized using a validated laboratory developed test (LDT) by one of CMPT’s dedicated reference laboratories- the measurand being a target’s endpoint PCR product by MALDI-TOF. These same samples also tested positive for these parasites by the BD Max Enteric Parasite Panel and were positive by initial microscopy. CMPT’s second reference laboratory ran the samples using the BioFire Film Array platform and confirmed expected results.  CMPT first sent the GIP PT samples in April followed by a second challenge round in August 2024 (Table 1).

Table 1. GIP PT scheme design

April 22, 2024		August, 2024
Sample ID	Organism	Sample ID	Organism
GIP2404-1	Salmonella species	GIP2408-1	Norovirus
GIP2404-2	Adenovirus	GIP2408-2	Negative (normal flora)
GIP2404-3	Giardia lamblia	GIP2408-3	Entamoeba histolytica
GIP2404-4	Negative (normal flora-E. coli)	GIP2408-4	Rotavirus
GIP2404-5	Yersinia enterocolitica	GIP2408-5	STEC (E. coli O157 H7)
GIP2404-6	Clostridioides difficile	GIP2408-6	Cyclospora species
GIP2404-7	Vibrio spp	GIP2408-7	Cryptosporidium species
GIP2404-8	Shigella species	GIP2408-8	Campylobacter species

Results

There were no issues detected in the first round of PT (GIP2404) with all participants receiving acceptable grades and all labs identified the pathogens correctly on their individual systems. During the second PT round (GIP2408) however, a difference in results was observed for two of the parasite organism targets: Entamoeba histolytica (GIP2408-3) and Cyclospora species (GIP2408-6).

When the results were analyzed and separated by test method, only the participants using the Biofire platform were able to detect the organisms while none of the other methods detected either of these two parasites in this challenge round (Table 2).

Table 2. PT results obtained by participants based on their testing platforms

Sample (target)	Reported	Biofire (7)	Seegene (4)	BD Max (1)	Other (1)
GIP-3 (E. histolytica)	E. histolytica	7
	B. hominis		3		1
	No pathogens		1	1
GIP-6 (Cyclospora spp.)	Cyclospora spp.	7
	No pathogens		4	1	1

These results prompted an investigation by CMPT to find out the possible cause(s) of these particular results.

Error Investigation

CMPT sent residual original neat stool and PT samples to BCCDC to be re-tested by various methods.

1. Entamoeba histolytica
   1. 1mL of neat stool
   2. 1mL of sample prepared as sent for PT testing
2. Cyclospora species
   1. 1mL of neat stool
   2. 1mL of sample prepared as sent for PT testing

Methods used:

1. Microscopy: wet mount, iron-hematoxilin (IH) stain, modified acid fast stain (mAB) stain (post-preservation of stool)
2. Antigen: E. histolytica antigen
3. PCR: BCCDC laboratory-developed qPCR

Results
Table 3. Summary of the results obtained by BCCDC when testing neat stool (NS) and PT samples (PT) by various methods

	Microscopy						E.h Ag	E.h/Ed qPCR		Cyclospora LDT qPCR
	NS			PT			NS	NS	PT	NS	PT
	Wm	IH	mAB	Wm	IH	mAB	-
E. histolytica	Neg	Neg		Neg	Neg		Neg	E.h and E.d detected	E.h and E.d detected
	NS			PT						NS	PT
Cyclospora spp.	Pos	Pos	Pos	Pos	Pos	Pos				Cyclo, detected	Cyclo. detected

NS: neat stool; PT: PT sample; Wm: wet mount; IH: Iron hematoxilin stain; mAB: modified acid stain; Ag: antigen; E.h: Entamoeba histolytica; E.d: Entamoeba dispar; LDT: laboratory developed test; Cyclo: Cyclospora; Neg: negative; Pos: positive.

E. histolytica investigation: The results indicate the sample that was identified as positive for E. histolytica was also positive for E. dispar.

Cross-reactivity between E. histolytica and E. dispar when using the BioFire has been reported by the manufacturer of this assay and as stated in the FilmArray Gastrointestinal (GI) Panel CE IVD Instruction Booklet (Biomerieux, RFIT-PRT-0143-07, August 2023[2])  “This assay may cross-react with the closely related E. dispar when present at higher levels (approximately 10⁵ oocysts/mL or greater)”.  Interestingly, in this case, the concentration of E. dispar appeared to be higher than E. histolytica when assessed by cycle threshold burden, and supported true and reproducible co-detection of these two organisms in the same sample. Lower organism burden, sample pre-treatment and 20x dilution of the original clinical sample may have led to the E. histolytica concentration falling under the limit of detection for some detection methods. Negative microscopy results suggest degradation of the sample over time.

[2] https://www.biofiredx.com/support/documents/#toggle-id-3 

Cyclospora spp.  investigation: The results also indicate that the sample identified as positive for Cyclospora spp. was confirmed as such by microscopy and LDT qCPR.  Laboratories employ different sample processing methods. Further investigation revealed that lack of bead beating led to a large (over 5 cycle threshold difference) compared to processing the same fecal swab sample with bead beating for Cyclospora spp., supporting the importance of sample pre-processing for appropriate detection of this organism.

Conclusions

Modern technologies have added another dimension to the microbiology laboratory allowing the simultaneous testing of multiple organisms in one sample. The implementation of these technologies needs to follow proper validation and optimization procedures and guided by experienced molecular technologists in order to make sure the assay is producing the results expected.

As demonstrated here, providers of PT services like CMPT are positioned to monitor and advise on matters relating to diagnostic test performance in the province. CMPT does not have access to nor can comment on the performance of other laboratories who participated in non-CMPT PT schemes, however enabling greater insight into the issue at the laboratory system- level is justified.

CMPT encourages laboratory stakeholders to share their PT results and to collaborate in the review of PT issues to prevent an escalation of error and ensure diagnostic accuracy in the spirit of quality patient care and treatment.

Acknowledgements

CMPT acknowledges the invaluable contributions of their members and clinical laboratory partner network across Canada which enables CMPT to fulfill its quality service mission. We are especially thankful to Dr. Catherine Hogan and the BCCDC Public Health Laboratory Parasitology and Molecular Microbiology and Genomics laboratories; Dr. Romina Reyes, Dr. Grace Yim and the LifeLabs technical team, and the CMPT Parasitology Technical committee members.

Respectfully at your service,

Canadian Microbiology Proficiency Testing Program (CMPT) 
Department of Pathology and Laboratory Medicine, The University of British Columbia

Vancouver, Canada
https://cmpt.ca/