Tag Archive for molecular

Validating molecular assay using R






Validation of Hepatitis B viral count using R-computing


Validation of Hepatitis B viral count using R-computing

This is an example on how R-computing can be used for validation of an quantitative assay. In this case two assays for Hepatitis B viral count are compared.

## Loading required package: seriation
## Loading required package: nlme

In a summary. 'Zero' values have been changed to '1' in order to be able to plot in logaritmic scale. The lower limit of detection (LLD) at home-lab is 10 IU/ml and the LLD at the reference-lab os 20 IU/ml. So, if the result is <20IU/ml, the detected value could be anywhere between 1 and 20. Therefore, the lower limit of detection has been set for home-lab at '5 IU/ml' and '10 IU/ml' for the reference lab.

summary(HepB_Web)
##       PIN              Ref_lab            Home_lab       
##  Min.   :14091022   Min.   :1.00e+00   Min.   :1.00e+00  
##  1st Qu.:14104055   1st Qu.:2.24e+02   1st Qu.:6.39e+02  
##  Median :14121724   Median :1.98e+03   Median :2.17e+03  
##  Mean   :14116291   Mean   :1.64e+07   Mean   :2.15e+07  
##  3rd Qu.:14132019   3rd Qu.:1.52e+05   3rd Qu.:8.42e+05  
##  Max.   :14132394   Max.   :1.70e+08   Max.   :2.88e+08
head(HepB_Web)
##        PIN Ref_lab Home_lab
## 1 14091022       1      184
## 2 14091023    3473     3473
## 3 14104024    2976     2558
## 4 14104025     988     1001
## 5 14104026   96670 20892951
## 6 14104141 1526000  1048129

To make it more easy, the set of values from Reference-lab = 'x'. The set of values from Home-lab = 'y'

Calculate the means and difference between the two sets (x and y)

# derive difference
mean(x)
## [1] 16447938
mean(y)
## [1] 21548265
# mean Ref_lab - mean Home_lab
mean(x)-mean(y)
## [1] -5100327

Because n=17 is small, the distribution of the differences should be approximately normal. Check using a boxplot and QQ plot. There is some skew.

HepB_Web$diff <- x-y
HepB_Web$diff
##  [1]       -183          0        418        -13  -20796281     477871
##  [7]         77   12039815       -176   34930655 -118402140       -282
## [13]         -9     -53972       -171      -1757          0        265
boxplot(HepB_Web$diff)

plot of chunk Boxplot

qqnorm(HepB_Web$diff)
qqline(HepB_Web$diff)

plot of chunk Boxplot
Shaphiro test of normality.

shapiro.test(HepB_Web$diff)
## 
##  Shapiro-Wilk normality test
## 
## data:  HepB_Web$diff
## W = 0.479, p-value = 5.294e-07

The normality test gives p < 0.003, which is small, so we
reject the null hypothesis that the values are distributed normally.

This means that we cannot use the student t-test. Instead, use the Mann-Whitney-Wilcoxon Test. We can decide whether the population distributions are identical without assuming them to follow the normal distribution.

wilcox.test(x, y, paired = TRUE)
## Warning: cannot compute exact p-value with zeroes
## 
##  Wilcoxon signed rank test with continuity correction
## 
## data:  x and y
## V = 59, p-value = 0.6603
## alternative hypothesis: true location shift is not equal to 0

p > 0.05 and therefore the H0 is NOT rejected.
The two populations are identical.

Just to see what happens in the Student T-test.
A paired t-test: one sample, two tests
H0 = no difference; H1 = mean of 2 tests are different
mu= a number indicating the true value of the mean
(or difference in means if you are performing a two sample test).

t.test(x, y, mu=0, paired=T, alternative="greater")
## 
##  Paired t-test
## 
## data:  x and y
## t = -0.7202, df = 17, p-value = 0.7594
## alternative hypothesis: true difference in means is greater than 0
## 95 percent confidence interval:
##  -17420746       Inf
## sample estimates:
## mean of the differences 
##                -5100327

p = 0.759. Because p is larger than alpha, we do NOT reject H0.
In other words, it is unlikely the observed agreements happened by chance.
However, because the populations do not have a normal distribution, we can not use the outcome if this test.

For correlation, three methods are used: pearson, kendall and spearman at a confidence level of 95%.

# correlation of the two methods
cor.test(x, y, 
         alternative = c("two.sided", "less", "greater"),
         method = c("pearson", "kendall", "spearman"),
         exact = NULL, conf.level = 0.95)
## 
##  Pearson's product-moment correlation
## 
## data:  x and y
## t = 11.19, df = 16, p-value = 5.646e-09
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  0.8472 0.9784
## sample estimates:
##    cor 
## 0.9416

The correlation with the spearman test is 0.9416175. Almost perfect correlation.

Plotting the two methods using logarithmic scales.

g <- ggplot(HepB_Web, aes(log(Home_lab), log(Ref_lab)))

# add layers
g + 
  geom_smooth(method="lm", se=TRUE, col="steelblue", size = 1) +
  geom_point(size = 3, aes(colour = x)) +
  scale_colour_gradient("IU/ml", high = "red", low = "blue", space = "Lab") +
  labs(y = "Reference lab (log IU/ml)") +
  labs(x = "Home lab (log IU/ml)") +
  theme_bw(base_family = "Helvetica", base_size = 14) +
  scale_x_continuous(breaks=c(0,4,8,12))

plot of chunk Plotting
Summary data on the correlation line.

regmod <- lm(y~x, data=HepB_Web)
summary(regmod)
## 
## Call:
## lm(formula = y ~ x, data = HepB_Web)
## 
## Residuals:
##       Min        1Q    Median        3Q       Max 
## -76044958   1901358   1905277   1905580  47898082 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -1.91e+06   5.98e+06   -0.32     0.75    
## x            1.43e+00   1.27e-01   11.19  5.6e-09 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 23800000 on 16 degrees of freedom
## Multiple R-squared:  0.887,  Adjusted R-squared:  0.88 
## F-statistic:  125 on 1 and 16 DF,  p-value: 5.65e-09

The Bland-Altman Analysis. To check if there is a bias.

##      Ref_lab  Home_lab       diff
## 1          1       184       -183
## 2       3473      3473          0
## 3       2976      2558        418
## 4        988      1001        -13
## 5      96670  20892951  -20796281
## 6    1526000   1048129     477871
## 7        919       842         77
## 8   23250000  11210185   12039815
## 9        421       597       -176
## 10 101000000  66069345   34930655
## 11 170000000 288402140 -118402140
## 12       483       765       -282
## 13         1        10         -9
## 14    169800    223772     -53972
## 15       158       329       -171
## 16        22      1779      -1757
## 17         1         1          0
## 18     10970     10705        265
BlandAltman(x, y,
            x.name = "Reference lab IU/ml",
            y.name = "Home lab IU/ml",
            maintit = "Bland-Altman plot for HBV count",
            cex = 1,
            pch = 16,
            col.points = "black",
            col.lines = "blue",
            limx = NULL,
            limy = NULL,
            ymax = NULL,
            eqax = FALSE,
            xlab = NULL,
            ylab = NULL,
            print = TRUE,
            reg.line = FALSE,
            digits = 2,
            mult = FALSE)
## NOTE:
##  'AB.plot' and 'BlandAltman' are deprecated,
##  and likely to disappear in a not too distant future,
##  use 'BA.plot' instead.

plot of chunk unnamed-chunk-6

## 
## Limits of agreement:
## Reference lab IU/ml - Home lab IU/ml                           2.5% limit 
##                             -5100327                            -65195645 
##                          97.5% limit                             SD(diff) 
##                             54994992                             30047659

When the dots are around 0, the two test could be interchanged for a patient. There are, however, some outliners: large difference of viral count between the two labs. One difference can be accounted for; the upper limited value of the reference lab is '>170.000.000 IU/ml, whereas the home-lab produces an exact calculation of 288402140 IU/ml.

Wish list for molecular lab

A clinical laboratory in Ethiopia wants to expand on molecular diagnostics. In order to get an idea what the ‘expanses’ are, I prepared a list of consumables, equipment, etc. Actually I was searching Internet if anyone is sharing such a list already, which wasn’t the case. So I prepared my own list which is available for anyone. The prices of all the items are mostly derived from the Fisher Scientific catalogue.

PCR rooms

The molecular lab consists of three physically separated rooms. In low resource settings, air control might be difficult. For PCR-1 it is important that air movement is limited to the minimum. For this, a PCR-cabinet can be used.

Here is the excel file with  the list of items for a molecular diagnostic laboratory. Have fun with it.

checklist PCR labo

In another excel file, I calculated exactly what the costs would be if, for example, the molecular diagnostic lab would perform an X number of multiplex real-time PCRs, for Y number patients, Z times a year. You can ask me if you need help in calculating the investment needed for your lab.

Training molecular diagnostics Module 1

Presentations of the Molecular Diagnostics Training in April 2014, Addis Abeba, Ethiopia.

Several pictures in the presentation were depicted from the web, publications and work from colleagues. I haven’t been able to ask permission from each of them, and I hope that no one will mind. The presentations are free to be shared for educational purposes. If you have any comments or objections, please let me know.

Module 1.1

Module 1.2

Module 1.3

Module 1.4

Module 1.5

Module 1.6

Training material

 

 

Molecular diagnostics in Ethiopia

Ethiopia was mentioned by the Washington Post as one of the top ten up coming economies in the world, however, extreme difficult business environment severely hamper growth aspects. This does not only apply for business, but for all entrepreneurs. From agricultural cooperatives to the health sector.

In Ethiopia there is a great demand for quality health services. The suffocating bureaucracy is hampering and damaging the sector. Crucial spare parts for diagnostic equipment are waiting in line between the shampoo and socks at he customs for weeks (months?). Human samples sent for analysis to Europe are withheld for obscure reasons. Well, let’s look at it from the optimistic point of view and suppose that regulations in Ethiopia will be more submissive in the near future.

In the medical microbiology a major development is taking place in which conventional diagnostics are being replaced and expanded by molecular biological tests. These tests are for a majority based on nucleic acid amplification. High sensitivity, a relative short turn-around-time in combination with an increasing reliability and robustness are characteristics for molecular diagnostics. Therefore, also in Ethiopia there is great interest for molecular diagnostics. Some laboratories have set up molecular tests,  for research or for diagnostics.

Molecular tests are being used for clinical diagnostics in resource-poor settings, for example tuberculosis detection using the GeneXpert point of care machines by Medecins Sans Frontieres. This kind of application of molecular tests is not sustainable and only to be used for emergency situations. For research, the veterinary section in Jimma University apparently has a well-equipped PCR lab. The lab cooperates with the University of Gent, Belgium lead by Professor Gryseelt. In Addis Abeba, the college of Health Sciences has much interest in teaching and applying molecular techniques. At the Institute for Plant and Livestock Research (ILRI) is a ready-to-use molecular lab. Unfortunately, consumables are hard to get. The Ethiopian Health & Nutrition Research Institute (EHNRI) in Addis Abeba, is cooperating with the CDC (Atlanta, USA) to set up a reference laboratory for the East African Region. Molecular biology would obviously play here a major role. It is not clear in what direction the Institute wants to move; do they want to focus on diagnostics? On Research? Or both? It’s a government organ and rumors are that foreign parties are being deterred from cooperation because of the obstructing bureaucracy. Subsequently, they turn to commercial laboratories. Commercial laboratories, such as the International Clinical Laboratories are building on foreign private investors and are therefore much more driven to get things done. Unfortunately, they also have to deal with silly laws and un-cooperating civil servants. Still, some of these laboratories already have experience with molecular diagnostics and have developed the appetite for more.

Cryptosporidium gp60 subtyping.

Cryptosporidium spp. are coccidian protozoan parasites that infect various vertebrate and invertebrate hosts. At least seven Cryptosporidium species have been associated with gastro-intestinal disease in humans: C. hominis, C. parvum, C. meleagridis, C felis, C. canis, C. suis and C. muris. Of these, C. hominis and C. parvum are the two species found most often in humans. Cryptosporidium-detection methods based on PCR has been an important instrument for studying the taxonomy and transmission of the parasite. For the differentiation of Cryptosporidium species/genotypes, different types of molecular tools have been used. In many studies the subtyping is performed using the Small SubUnit ribosomal RNA (SSU rRNA). Another genetic target is an oocyst wall protein (COWP) gene. However, the COWP gene is less useful because of its narrow specificity (Xiao, 2010). One gene that is becoming popular for subtyping Cryptosporidium is the 60 kDa glycoprotein (gp60). A Cryptosporidium subtype will end up with a name such as “IaA23R4” or “IIdA18G1”. In this blog I will explain the methodology behind this classification of Cryptosporidum subtypes based on gp60 analysis.

The gp60 gene has tandem repeats of the serine-coding trinucleotide: TCA, TCG or TCT. These tandem repeats have varying lengths. The DNA sequence outside these repeats also shows considerable differences; they are used to categorize C. parvum and C. hominis each into several subtype families. The name of gp60 subtypes start by designating “I” for C. hominis and “II” for C. parvum. Subsequently, each new subtype family is given a letter. Ia, Ib, Id, etc. for C. hominis and IIa, IIb, IIc, etc. for C. parvum. What follows then is the letter and number of the trinucleotide repeats. TCA is represented by the letter A, TCG is represented by the letter G and TCT by the letter T. The letter R is designated for any other repeat.

So lets look at some examples. In this text-file you may find some Cryptosporidium subtype DNA sequences that I used: three C. hominis subtypes and three C. parvum subtypes. 

cryp alignment Screen Shot 2013-08-13 at 22.27.46

By the way, here are two useful online tools for DNA sequence analyses, which I often use. For DNA aligment: Multalin …and to draw phylogenetic trees: ClustalW2

For the first example I took the DNA sequence with GenBank Accession number AF164502.

Cryp subtype1.001

The dominant trinucleotide repeat is TCA (in red). The serine trinucleotide is repeated 23 times. Another repeat is with AAGACGGTGGTAAGG (in green). The sequence is repeated three times. However, the repeat is followed by a DNA sequence similar to the preceding one: AAACGGTGAAGG. In this case the researcher has decided that it should be part of the tandem repeat. So here we end up with a ‘Cryptosporidium hominis’ of the subtype family ‘Ia’ followed by ‘A23R4’.

Another one. Let’s take Accession number AY262034, C. parvum subtype family IIa.

Cryp subtype2.002

The dominant trinucleotide is TCA with 15 repeats (purple). In between are two TCG’s (green). Furthermore, this subtype also has one ‘R’ (other) repeat: ACATCA (orange). Huh? One repeat? Well, apparently there are similar subtypes with two copies of ACATCA. Hence, R=1 in this one, which makes IIaA15G2R1.

Ok, last one. Genbank number AF164491. A C. parvum beloning to subtype family IIc.

Cryp subtype3.003

Five times TCA (blue) and three times TCG (pink). In the IIc subtype family, all strains have A5G3 repeats. However, downstream in the 3’ region some strains are different. The diverged strains have subsequently an extension. In this case it’s IIcA5G3a.

Subtyping microbiological strains always had a slice of guesswork. How many nucleotide substitutions does it need to make the strain to be a new genotype? When would you denote a sequence to be a repeat? Molecular genotyping has provided many advantages over conventional typing methods. Sometimes it might seem that the resolution for differentiating strains with molecular methods are either to narrow or to broad. Still, as we see with the typing for Cryptosporidium strains, it takes a lot of effort, thought slowly a consensus is appearing.

Want to know more about sub typing Cryptosporidum strain? A good start is the article ‘Molecular epidemiology of cryptosporidiosis: An update‘ by Lihua Xiao in Experimental Parasitology (124 (2010) 80-90). Or if you have any specific questions, feel free to contact me.

Chapter 11 – General Discussion

Molecular diagnosis of intestinal parasites

Real-time PCR has been established as a sensitive and specific qualitative and quantitative technique within the fields of routine diagnostic virology and bacteriology (Read et al., 2000; Niesters, 2002; Klein, 2002; Mackay, 2004; Iijima et al., 2004; Espy et al., 2006; Schuurman et al., 2007a)⁠. Several real-time PCRs for the detection of intestinal parasites have been developed showing excellent specificity and sensitivity and accepted as objective tools for case confirmation and gold standard in the development of new diagnostic tools. Implementation of molecular methods in routine diagnostics moved a step forward with the development of the first multiplex real-time PCR for the simultaneous detection of the three most important diarrhoea causing protozoa (Verweij, 2004)⁠. Recent developments in improved and simplified DNA isolation and automatizing of these methods and the broader availability of real-time PCR in diagnostic microbiology laboratories will allow these methods to be used as an alternative tool for the routine diagnosis of intestinal parasitic infections (Verweij, 2004; Espy et al., 2006)⁠. The high specificity of DNA-based methods is one of major advantages and at the same time one of the major constraints as these methods will only detect the specific DNA targets for which primers and detection probes are designed. This is in contrast with the “overall view” of conventional diagnosis of intestinal parasites using microscopy. This thesis describes a number of studies on the implementation of real-time PCR for the detection and quantification of intestinal parasites to validate and optimize it’s use in different diagnostic settings and in epidemiological surveys.

Sample collection

In industrialized countries it is common practice that stool samples are sent to the laboratory by regular mail. This delay between stool production and analysis has little or no effect on the microscopic detection of parasite eggs and cysts. The same applies for the detection of parasitic DNA with real-time PCR; in a study on the DNA detection of Dientamoeba fragilis, a parasite which lacks a cyst stage, isolation and the outcome of the PCR was reproducible even after 8 weeks of storage at 4°C (Verweij et al., 2007b)⁠.

For epidemiological studies, the storage of stool samples might nevertheless be a major problem. It is not exceptional that these studies are conducted in remote areas with a (sub)tropical climate, where power cuts are a daily phenomenon. To avoid degradation of collected samples, faecal specimens can be suspended in ethanol which will preserve the samples for at least 3 months even at tropical temperatures. This is illustrated by the detection of Schistosoma haematobium DNA in stool samples collected from 10 infected persons (figure 11.1). Although S. haematobium eggs are normally excreted via the urine, in chapter 9 S. haematobium DNA detection in stool is shown to be feasible. Figure 11.1a shows similar DNA amplifications of isolated S. haematobium DNA from stool samples that were stored either frozen or suspended in ethanol at room temperature. Storage and transport of non-fixated stool samples at room temperature as a control was not feasible under the tropical conditions. For urine, S. haematobium DNA was detected in most samples stored at different temperatures (figure 11.1b). However, the temperature at which the urine samples were stored does influence the quantitative results.

Chapter 10 describes a sensitive method for the detection of S. haematobium DNA from vaginal washings. These samples were collected under difficult conditions and at that time the researchers postulated that a more simple collection procedure and diagnostics are necessary. The use of vaginal swabs already proved to be a simple and sensitive method for the detection of several female genital infections (Whiley et al., 2005; Caliendo et al., 2005; Jaton et al., 2006)⁠. It is therefore worthwhile to investigate if vaginal swabs could also be applied for the diagnosis of female genital schistosomiasis. For epidemiological studies, reliable storage methods which preferably allow the intensity of infection to be (semi-)quantified, still need to be tested both for urine samples and vaginal swabs.

 

Figure 11.1. Detected Schistosoma haematobium Cycle threshold (Ct)-values of DNA isolated from stool samples (a) collected from 10 patients living in an endemic area with S. haematonium infection determined by microscopic egg counts in urine. Samples of stool were stored in frozen condition or as ethanol suspension at room temperature over a period of two weeks. Additionally, urine samples (b) were collected and divided in three parts, with each part stored at a different temperature for a period of two weeks. No ethanol or other fixatives were added to the urine. S. haematobium Ct-values were detected from DNA isolated from the urine samples. Urine and stool samples were collected and kindly provided by Dr. Akim Adegnika in collaboration with the medical research unit of the Albert Schweitzer Hospital in Lambaréné, Gabon.

11.1a 11_1b

 

Sample analysis

The isolation of parasitic DNA is one of the most important steps for the implementation of DNA-based methods. It is therefore essential to check for each new target if the isolation procedure is capable of extracting the DNA from the faecal sample. For example, a Trichuris trichiura-specific PCR was successful for the detection of T. trichiura-DNA extracted from an adult worm, however, isolation of DNA from the T. trichiura eggs in faeces failed. Several pernicious treatments (e.g. sonification and microwaving) have been used in vain to release the DNA from the eggs. DNA isolation from T. trichiura eggs remains a challenge, even more so if considered that the method should be feasible for routine sample preparation.

Faecal contaminants can cause inhibition of the DNA amplification. An internal control is therefore essential to monitor the efficiency of the amplification process. One of the targets of a multiplex assay amplifies and detects Phocin Herpes virus (PhHV) (Niesters, 2002)⁠, of which a fixed amount is added to one of the isolation buffers. A higher Ct-value than expected or the absence of the signal indicates a less efficient amplification process which can result in a false negative outcome of one of the other targets. PhHV is preferred because of simplicity and standardization: the same buffer with the internal control is also used for all multiplex assays and for other samples, such as tissue or urine samples. Detection of PhHV amplification does not necessarily mean that faecal DNA has been isolated. DNA amplification from normal flora found in the gastro-intestinal track has been suggested as control for faecal DNA isolation. However, the amount of normal gut flora fluctuates between samples resulting in different Ct-values and as a consequence (partial) inhibition would not be recognized. Another suggestion is to include one positive faecal sample for each target within a series of samples prepared for DNA isolation. This procedure would need sufficient supply of positive material and it still can not assure that DNA isolation in other samples has succeeded.

Real-time PCR offers the ability to amplify and detect multiple DNA targets in a single reaction. Important advantages of a multiplex assay include reduced reagent costs and a high throughput potential. However, during the DNA amplification, inhibition can occur due to the simultaneous amplification of multiple targets in one assay. Part of designing a multiplex real-time PCR involves a background test, in which the amplification of one target is measured in the presence and absence of a second target. The background test shows the performance of the assay to detect a small number of DNA copies of one target in the presence of more abundant copy numbers of another parasite species. Combining of the real-time PCR for the detection of Isospora belli detection (Ten Hove et al., 2008)⁠ with the microsporidia real-time PCR (Verweij et al., 2007c)⁠ into one assay failed at this step. Amplification of multiple targets was not possible without the inhibitive effect on the amplification of the microsporidia DNA. It is not clear yet if the performance of this multiplex assay will improve by using another type of real-time PCR apparatus. Otherwise, a new design of these assays is needed.

Several real-time PCRs are already effectively employed in routine diagnosis and epidemiology. Because of the high specificity of real-time PCR, an assay could fail to detect genetically related species or strains (e.g. Cryptosporidium species and E. histolytica strains) due to variations in the DNA sequence of the primer and probe regions. Also, rare parasite species will be missed with standard real-time PCR panels. In patients with continuing gastro-intestinal complaints in which the initial real-time PCR screening does not reveal the cause of the complaints, microscopy is still required as an additional tool. The challenge for the future will be to maintain a profound knowledge on the morphological characteristics of in these more difficult academic cases.

In routine diagnostics and in epidemiology, often large numbers of samples are processed. With increasing numbers of samples, the amount of hands-on time for sample preparation and analysis can become a major bottleneck and human errors might occur more easily. Microscopy can not be subjected to automated process whereas molecular diagnostics can be incorporated into a high throughput robotic workstation. Several automated nucleic acid extraction platforms are available for various specimens and applications (Jongerius et al., 2000; Kessler et al., 2001; Espy et al., 2001; Knepp et al., 2003)⁠. Beside the reduction on hands-on time, increased consistency and reproducibility of DNA-extraction and -analysis will improve standard operation procedures in the laboratory.

Giardia lamblia

Giardia lamblia is worldwide one of the most common intestinal parasites causing gastro-intestinal complaints. Real-time PCR for this parasite is therefore considered as the first choice as a target in a diagnostic panel. Technical evaluation of the PCR already demonstrated 100% specificity and a higher sensitivity as compared to microscopy and antigen tests (Verweij et al., 2003b, 2004a)⁠. Table 11.1 gives the results of the diagnostic tests for G. lamblia in a routine academic setting (Clinical Laboratory for Microbiology, Leiden University Medical Centre). It is clear that real-time PCR shows a higher sensitivity for the detection of G. lamblia. In this thesis, G. lamblia real-time PCR was evaluated as part of a routine diagnosis for intestinal parasites in a western European setting. In previous studies, G. lamblia has been reported to be a common pathogen in The Netherlands with a prevalence of 5% among general practice patients with gastro-intestinal complaints (De Wit et al., 2001c)⁠. A comparable prevalence is reported in chapter 2 after routine screening with microscopy of general practice patients. When real-time PCR was applied on the same samples, the prevalence increased to 9.3%. In cases for which no microscopical examination for intestinal parasites was requested by the general practitioners, G. lamblia was detected in 6.5% of the samples. This finding concurs with those from two other studies (Mank et al., 1995a; Van den Brandhof et al., 2006)⁠. Mank et al. reported that more than a third of detected pathogenic protozoa, mainly G. lamblia, would have been missed if the laboratory would have complied with the general practitioners’ requests and Van den Brandhof et al. concluded that test request by general practitioners do not always comply with existing knowledge of the etiology of gastro-enteritis. Another reason for missed infections mentioned in the latter study, is the general restrictive policy in The Netherlands with regard to requesting tests.

Table 11.1. Test results (- = negative and + = positive) retrieved from the validation report for implementation of real-time PCR in routine diagnosis of Giardia lamblia at the clinical laboratory for microbiology at Leiden University Medical Centre (Unpublished data from Dr. J.J. Verweij).

Microscopy

Antigentest a

Real-time PCR

Number of samples

Ct range

Median Ct

254

+

+

+

75

20,1 – 38,0

26,1

+

+

9

27,3 – 36,7

32,0

+

24b

30,7 – 42,8

35,8

362

a Alexon-Trend ProSpecT EIA for the detection of G. lamblia.

b These samples include: 10 samples from four patients where in successive samples G. lamblia was also detected by microscopy and/or antigen test; 11 samples from six patients where G. lamblia infection was found in a family member; 3 samples from three patients in which only one out of a series of three was positive with PCR.

 

Although G. lamblia is regarded as a pathogenic parasite, in case-control studies it was also found in controls (Whitty et al., 2000; De Wit et al., 2001c; Hörman et al., 2004; Amar et al., 2007)⁠. In chapter 4, half of all detected cases of G. lamblia originated from returned travellers without gastro-intestinal complaints. The G. lamblia positive samples were further characterized in chapter 5 to analyse the role of G. lamblia assemblages (i.e. defined group of genotypes) in symptomatic and asymptomatic returned travellers. The data showed that increasing G. lamblia Ct-values is in concordance with decreasing gastro-intestinal complaints, translated as symptoms-score. However, the assemblages of G. lamblia revealed no association with the presence or absence of any clinical symptoms in this group of travellers. Previous studies suggested the epidemiological role of G. lamblia assemblages in children as an important factor associated with gastro-intestinal symptoms (Read et al., 2002; Haque et al., 2005; Sahagun et al., 2008; Kohli et al., 2008; Peréz Cordón et al., 2008)⁠. In adults, like most subjects in chapter 4 and 5, the association of assemblages with gastro-intestinal symptoms is less evident and can even give contradicting results (Homan & Mank, 2001; Gelanew et al., 2007; Sahagun et al., 2008)⁠. It is suggested that the presence of symptoms in adults more likely to be related with underlying conditions such as the degree of host adaptation, nutritional and immunological status. The results of studies on the association of G. lamblia assemblages with gastro-intestinal complaints are summarized in table 11.2.

 

Table 11.2. Summary of studies on the association of Giardia lamblia assemblages with clinical symptoms.

       

Clinical symptoms

Reference Study area Study group Gene target

Assemblage A

Assemblage B

Homan et al., 2001 Netherlands n=18; 8-60yrs GDH

Mild, intermittent diarrhoea

Severe, persistent diarrhoea

Read et al., 2002 Australia n=23; <5yrs SSU-rDNA

Symptomatic

Asymptomatic

Haque et al., 2005 Bangladesh n=211; <10yrs SSU-rDNA

Symptomatic

Asymptomatic

Gelanew et al., 2007 Ethiopia n=80; all ages β-giardin

Less symptomatic

More symptomatic

Sahagun et al., 2008 Spain n=72; <5yrs TPI

More symptomatic

Less symptomatic

n=65; 5-72yrs TPI

No difference

Peréz et al., 2008 Peru n=201; 0-9yrs GDH

Diarrhoeic

High shedding, non diarrhoeic

Kohli et al., 2008 Brazil n=189; children SSU-rRNA

No difference

Ten Hove et al., 2008 Worldwide

N=156; mostly adults

TPI

No difference

 

Future studies on transmission patterns in households in combinations with detailed description of underlying conditions could give more inside on the role of G. lamblia assemblages on the clinical presentation.

Cryptosporidium

Cryptosporidium infections were studied in both the general practice patients and returned travellers. Microscopic detection of Cryptosporidium requires additional staining methods to visualize the parasite in faecal samples. This additional diagnostic procedure is not always included in routine stool analysis or requested by the health care providers, probably because cryptosporidiosis is usually associated with immune-depression, in particular HIV/AIDS. Given the sporadic requests from the health care providers for additional Cryptosporidium detection, prevalence of cryptosporidiosis is highly underestimated (chapter 2) (Van Gool et al., 2003; Jones et al., 2004)⁠. The prevalence of Cryptosporidium in patients with diarrhoea as described in chapter 2 and 3 has so far exceeded the figures provided in other studies (Petry, 1998; Banffer & Duifhuis, 1989; Rodriguez-Hernandez et al., 1996; De Wit et al., 2001a; Vandenberg et al., 2006a). In the month of September 2005, Cryptosporidium (mainly C. hominis) was detected in up to 30% of all children aged under five with gastro-intestinal complaints. The reason for the high prevalence described in these chapters could be due to the unselected screening of patients by highly sensitive real-time PCR analysis in combination with sample collection during a seasonal peak of cryptosporidiosis (Van Asperen et al., 1996; Wielinga et al., 2007)⁠. A common source of infection has not been identified but cases did not show any geographic clustering (unpublished observations). A low prevalence of Cryptosporidium was shown in the group of mainly adult travellers, even among those who returned from ‘high risk areas’.

Entamoeba histolytica

The very low prevalence of Entamoeba histolytica infections in the general practice patient population of chapter 2 could suggest that the E. histolytica assay is redundant in the diagnostic real-time PCR panel. Although infection with E. histolytica is rare in The Netherlands, the assay was included because of its major clinical importance for the patient and its potential to spread among household members (Stanley, 2003)⁠. Even though a clinical E. histolytica infection is normally associated with tropical areas, this is not always evident and can therefore delay the diagnosis (Vreden et al., 2000; Edeling et al., 2004; Veneman et al., 2006)⁠. Such a case emerged during screening for Cryptosporidium with the E. histolytica-G. lamblia-Cryptosporidium PCR (HGC-PCR) of an additional 1000 general practice patients as described in chapter 3. The E. histolytica real-time PCR positive individual concerned a young male with severe diarrhoea and an unknown travel history. Despite several microscopic analyses over a period of more than 3 years, still no pathogen was detected that could explain the gastro-intestinal complaints. Eventually the patient was treated as a clinical amoebic infection (T. Schuurman, personal communication). As described in the introduction, the classical method for detection and diagnosis of E. histolytica infection is not that straightforward and it is essential that physicians are familiar with epidemiology, clinical syndromes, management of infections and available diagnostic tests. The introduction of the HGC-PCR, in particular for travellers, greatly facilitates the difficult diagnosis of E. histolytica infections.

Strongyloides stercoralis

The diagnosis of a Strongyloides stercoralis infection by microscopy involves multiple sampling and concentration methods. The Baermann concentration method is widely used, but the sensitivity of the test relies on the number of samples examined. Furthermore, there are several reasons why this S. stercoralis detection procedure is omitted, like in chapter 4 because the samples were too old, did not have the right consistency, or only a small amount of faeces was available. Moreover, it was shown that physicians often don’t consider strongyloidiasis in their differential diagnosis, which could place patients at iatrogenic risk (Loutfy et al., 2002; Boulware et al., 2007)⁠. In the setting described in chapter 4, the screening for S. stercoralis infection with real-time PCR proved rewarding, but may also be an important method to avoid hyper-infection in patients in need of immune-suppressive treatment.

Isospora belli and microsporidia

In the epidemiology of opportunistic intestinal parasitic infections in immune-suppressed and immune-incompetent patients, real-time PCR proved to be a useful tool for the detection of Isospora belli (chapter 6) and microsporidia (chapter 7) infections. In Blantyre, Malawi, the real-time PCRs were used for a study on the aetiology of diarrhoea in hospitalized patients (figure 11.2). Stool samples were collected from HIV/AIDS positive- (n = 226) and HIV-negative- (n = 46) patients. I. belli and Enterocytozoon bieneusi were detected by real-time PCR in 175 (64%) and 197 (73%) of all cases, respectively. Preliminary data analyses show that the intensities of the parasitic infections (reflected in Ct-values) were associated with the severity of diarrhoea (Beadsworth, unpublished data). Moreover, I. belli was more common in patients with 100-200 CD4-cells/mm3, while E. bieneusi was mostly detected in patients with <100 CD4 cells/mm3 (figure 11.2).

11_2


Figure 11.2. Percentage of detected Isospora belli positive cases and Enterocytozoon bieneusi cases with detected Ct-values below 35, in HIV-negative (n=46) and -positive (n = 226) patients categorized by their HIV status (reflected by CD4 counts). Samples were collected from patients in Blantyre, Malawi (Unpublished data from L. van Lieshout, M.B. Beadsworth et al.).

 

The decrease in the number of I. belli infections in the last patient group is not necessarily the result of the lower CD4 counts (Kartalija & Sande, 1999; Certad et al., 2003)⁠. It could also be due to the in vivo competition between the two parasitic infections. Further data analysis, including the results of Cryptosporidium detection is still pending (Beadsworth et al., in preparation).

Because of higher awareness of microsporidiosis together with the availability of the highly sensitive and specific diagnostic techniques based on PCR, microsporidia infections are increasingly diagnosed in transplant recipients, children, seniors and travellers. The detection of microsporidia infections in these patient populations in western settings is mostly clear cut. However, in the study in Malawi, but also in unpublished observations in Ethiopia, a very high prevalence of E. bieneusi infections is observed. In half of the detected cases Ct-values were above 35 reflecting a very low DNA-load. The clinical relevance of these findings is not clear. Possible explanations for these cases could be just an infection with a low parasite load, the presence of asymptomatic carriers or the results of an ubiquity of spores in tropical environments with high HIV/AIDS prevalence among the population (Van Gool et al., 1997; Cegielski et al., 1999; Tumwine et al., 2002; Siński, 2003; Nkinin et al., 2007; Samie et al., 2007; Raccurt et al., 2008)⁠. To elucidate the association of microsporidia infections with different clinical backgrounds, E. bieneusi genotypes were characterised in several patient cohorts. The phylogenetic study in chapter 8 describes a clustering in different human populations for some of the genotypes of E. bieneusi. Using the same techniques for detection and characterization, future studies could attempt to follow transmission patterns involving (asymptomatic) household members, domestic animals and the environment to determine the potential of E. bieneusi as a zoonotic and water-borne pathogen.

Schistosoma spp.

In chapter 9 a multiplex real-time PCR approach on the cytochrome c oxidase subunit 1 (cox1) gene was compared with egg counts determined by Kato-smears on preselected series of faecal samples. These were collected from subjects living in an area endemic for both Schistosoma mansoni and Schistosoma haematobium. The multiplex real-time PCR showed 100% specificity using an extensive number and variety of controls and 100% sensitivity in subjects with more than 100 eggs per gram faeces, with few false negatives in patients excreting lower numbers of eggs. Although eggs of S. haematobium were not observed in stool samples, its DNA could successfully be detected. As expected the sensitivity of S. haematobium DNA detection in faeces is lower than the sensitivity achieved by microscopic examination of urine samples. Preliminary results using real-time PCR on DNA isolated from urine samples, collected in a S. haematobium-endemic area in Gabon showed promising results (figure 11.1). For further improvement on the sensitivity of the molecular test, another real-time PCR was developed targeting the internal transcribed spacer 2 (ITS-2) (Obeng et al., 2008)⁠. The ITS-2 PCR improved Schistosoma detection considerably, in particular for the detection of S. mansoni (table 11.3). Unlike the cox1 gene, the real-time PCR assay on ITS-2 does not discriminate between DNA of S. mansoni and DNA of S. haematobium.

Table 11.3. Amplification of a specified amount of Schistosoma mansoni DNA and S. haematobium DNA isolated from adult worms, with the species specific cox1 real-time PCR assay and the genus specific ITS-2 real-time PCR assay reported in detected Ct-values.

 

 

Schistosoma mansoni

 

Schistosoma haematobium

Target DNA

 

cox1 PCR

ITS-2 PCR

 

cox1 PCR

ITS-2 PCR

100 ng

18,0

14,8

17,2

16,3

10 ng

21,8

17,2

20,5

19,5

1 ng

25,9

20,9

24,1

23,1

100 pg

29,9

24,4

27,7

26,7

10 pg

33,6

27,9

31,3

29,0

1 pg

37,6

31,2

33,8

34,4

100 fg

41,9

36,0

39,0

38,5

10 fg

0

0

0

0

 

The ITS-2 PCR was further evaluated using urine samples collected in Ghana and compared with CCA strips and microscopy (Obeng et al., 2008)⁠. The results indicate that real-time PCR may potentially serve as a gold standard to determine the prevalence and intensity of Schistosoma infections in surveys. Moreover, these results and the findings described in chapter 9 suggest the versatility of real-time PCR as a diagnostic tool for urine, stool and other clinical samples. In chapter 10 the performance of Schistosoma real-time PCR was evaluated for the diagnosis of female genital schistosomiasis on DNA isolated from vaginal lavages. This provisional investigation showed that Schistosoma real-time PCR is a valuable tool for diagnosis in research and evaluation projects.

multiplex real-time PCR for routine diagnostics

The implementation of real-time PCR in routine diagnostics has encountered several paradigms in the medical community on assumptions, concepts, practices and even values that constitute the way on how parasites should be diagnosed. The microscopic detection of parasites is generally accepted as the gold standard, although it has also been widely acknowledged that it falls short on effectiveness in both patient care and epidemiology.

Each year approximately 320.000 Dutch patients with gastro-intestinal complaints consult their general practitioner (Van Duynhoven et al., 2005)⁠. Guidelines for general practitioners to request stools were created by the Dutch Society of General practitioners (NHG) (Brühl et al., 2007)⁠ to improve daily practice for managing patients with gastro-intestinal complaints and can be summarized as follows:

The guidelines for the diagnostic standard for acute diarrhoea recommends additional diagnostic analysis for intestinal protozoa when the diarrhoea continues for more than 10 days. The GP should ask for a fresh stool sample for it to be analysed within hours after production. When the outcome is negative, two more stool samples should be requested, produced with an interval of a few days. If available, preference should be given to the TFT procedure. If no intestinal parasites were discovered in three samples and protozoan infection is still suspected, the laboratory could be consulted to perform a copro-antigen test for G. lamblia detection. Detection of E. histolytica with PCR and Cryptosporidium with specific staining methods have to be explicitly asked for as these techniques are usually not performed on routine basis. (Translated from Dutch by R. ten Hove).

 

Although the diagnostic standard procedure for general practitioners as described above is fairly straightforward, many additional patients visiting their general practitioner were diagnosed with G. lamblia- and Cryptosporidium-infection by the molecular diagnostic approach (chapter 2 and chapter 3). Detection of these cases with the conventional diagnostic approach had failed because cases did not met the criteria for parasite examination or, even if cases answered the criteria for parasite examination, specific parasite infections were not suspected by the general practitioners. Last but not least, supported by the semi-quantitative results, real-time PCR has the ability to detect very low loads of intestinal parasites that would most likely be missed with microscopic and antigen detection methods.

The design of real-time PCR panels for specific patient groups could improve the standard diagnostic procedures and patient diagnostics (Verweij, 2004)⁠. The advantages of such real-time PCR panels can further be supported by the results of chapters 2, 3 and 4. Table 11.4 gives examples of real-time PCR panels that could be used in different patient groups. In children, G. lamblia and Cryptosporidium are two important pathogens. Although D. fragilis has been suggested to be also a potential pathogen in children, more studies are needed to support this statement. Some of the microbiological diagnostic laboratories in The Netherlands have replaced microscopic examination of SAF preserved stool samples with real-time PCR panel including D. fragilis-specific PCR which can help to elucidate the pathogenic role of this parasite (Bruijnesteijn van Coppenraet et al., n.d.)⁠.

In immune-compromised patients the most important opportunistic intestinal parasites can be related to the conditions of the immune suppression. For example, S. stercoralis hyper infection is associated with corticosteroid treatment and HTLV infection but not with HIV infection.

Table 11.4. Real-time PCR assays that were developed and evaluated summarized as components of different patient groups. Several targets have also been combined and evaluated as multiplex real-time PCR assays.

AdultsEntamoeba histolytica

Giardia lamblia

Cryptosporidium

ChildrenDientamoeba fragilis

Giardia lamblia

Cryptosporidium

TravellersEntamoeba histolytica

Giardia lamblia

Cryptosporidium

Strongyloides stercoralis

Schistosoma b

Cyclospora cayetanensis b

Isospora belli b

E. bieneusi c

Immune compromisedStrongyloides stercoralis a

Cryptosporidium

Isospora belli

Enterocytozoon bieneusi

Encephalitozoon spp.

a: Recommended when corticosteroid treatment is scheduled.b: Optional for those returning from high-risk areas.

c: Actual prevalence among travellers needs further investigation.

 

Travellers comprise a very diverse group, including immigrants, adoption children, tourists, businessmen, expats, army personal, etc. that may have been exposed to a wide spectrum of infections. Many studies have assessed the relative frequency of intestinal parasitic diseases to characterize demographic and travel-related predictors of infections (Bottieau et al., 2006, 2007; Van De Winkel et al., 2007; Cobelens et al., 1998; Muennig et al., 1999; Von Sonnenburg et al., 2000; Whitty et al., 2000; Saiman et al., 2001; Okhuysen, 2001, 2007; Ansart et al., 2005; Steffen, 2005; Bailey et al., 2006; Thors et al., 2006; Freedman et al., 2006; Seybolt et al., 2006; Caruana et al., 2006; Wilson et al., 2007; Sudarshi et al., 2003; Boggild et al., 2006; Brouwer et al., 1999)⁠. To reach a consensus on which diagnostic procedures to perform for this patient group remains very complicated. In chapter 4 the overall number of detected cases using conventional- and the molecular-diagnostic approach was small. However, in the setting of a travel clinic the full arsenal of diagnostic tests are used to cover most of the intestinal pathogenic parasite species. Profit can be gained in simplifying these diagnostic procedures. An automated diagnostic screening system for the most important parasites can be more cost-efficient. Nevertheless, it is still difficult to decide which parasite targets are needed in the standard screening panel. Real-time PCR showed more effective in detecting all four targeted parasite species than with conventional diagnostic methods. Moreover, only few additional parasite species were diagnosed with microscopy, while many more unexplained causes of gastro-intestinal complaints might be resolved by the application of additional real-time PCR targets.

concluding remarks

Real-time PCR has opened new perspectives on diagnosis of intestinal parasites. Nevertheless, the process of improving diagnostic methods is still continuing. It is evident that in the future the throughput of real-time PCR platforms will improve substantially by increasing the number of targets, speeding up sample processing time and introducing plates with a larger number of wells. Already a real-time PCR system has been assessed for the detection of five viral targets on 384 well format (Molenkamp et al., 2007)⁠. Moreover, diagnostic technologies, such as biosensors, micro-DNA arrays and microfluid technologies are advancing (Wang et al., 2004; Van Merkerk, 2005; Taguchi et al., 2005; Stein et al., 2006)⁠. At the same time, new parasitic diseases can emerge while others disappear. Therefore, any diagnostic method, new or old, should only be applied when it has proven to provide meaningful analysis. The present thesis shows that real-time PCR provides a feasible and effective alternative method in the routine diagnosis of intestinal parasites in patient care and epidemiology.

The initial implementation of real-time PCR in a laboratory requires a large investment. On the other hand prices of the equipment and consumables continue to decrease and become more competitive compared with the high costs of staff in a laboratory. Cost-effectiveness of real-time PCR could increase when diagnostic assays are extended with additional parasitic, bacterial, fungal and/or viral targets. Nevertheless, it needs to be taken into consideration that although cost efficiency on consumables and labour-time might favour one diagnostic methodology, the amount of reimbursement for a laboratory can still vary extensively due to different declaration settlements in health care systems.

A major challenge that remains is the access to sensitive and specific diagnostic tools in low income countries. Although the real-time PCR method brings forward new exciting prospects on the epidemiology of intestinal parasites in remote and endemic areas, it has little value for clinical diagnostics in most of these settings. Real-time PCR can perform a prominent role as a gold standard in quality control for the evaluation of new inexpensive rapid tests for field diagnostics. Even more, field applicable PCR formats can be developed with instruments being further miniaturized, integrated into all sample processing and made less expensive (Gerdes et al., 2008; Yager et al., 2008)⁠. A giant leap in public health can be made if medical practices in resource poor settings could lean on such advances.

Multiplex sequence detection by Surface-enhanced resonance Raman scattering (SERRS)

Multiplex real-time PCR is nowadays routinely used in diagnostics. A multiplex real-time PCR assay targets multiple genes, with a gene-specific probe labeled with a different fluorophore. Each independend fluorophore takes a slice out of the light spectrum. These channels are carefully selected and tested to avoid any crosstalk between the channels. Nonetheless, it is sometimes unavoidable that one ‘channel shines through to the other’. The variables that affect the performance of an assay depend on the ingredients of the reaction, such as enzymes, primers, inhibitors and also the used fluorephores.

EmsFiltersFluorophores

The process of setting up an assay in which multiple targets in a mixture are detected can be tiring. Even when all chemicals for the reaction are optimized on a platform, still, spectral overlap can suddenly occur without a clear reason. Multiplex real-time PCR based on a light spectral detection platform is limited to five or six channels (or targets). In order to increase the number of DNA targets, a different format is desirable.

In the past few years a new technique has emerged. One in which both fluorophores and nonfluorophores (chromophores) are effectively discriminated, allowing a much more extensive multiplex reaction. This technique is called Surface-enhanced resonance Raman scattering (SERRS).

In a nut-shell how does Multiplex sequence detection by SERRS work? To illustrate the process, I will use a figure from the article of Cécile Feuillie et al. in PLoS May 2011.

SERRS PCR

Step 1. A multiplex PCR is performed using ordinary primers. Amplification products are captures by a chromophore-labeled probes (so not necessarily fluorescent labels). The hybridized structures are captured by a second Biotin labeled probe. The PCR product together with the two probes are immobilized by the binding of Biotin with Streptavidin-coated magnetic beads.  Then, unbound compounds are washed away. The probes are released and the chomophore-labeled probe is than available to be processed for SERRS measurements.

Step 2 involves the detection of SERRS active probes (Biotin, by the way, is not SERRS active). One important aspect in SERRS is that the detection is strongly enhanced when the probe is adsorbed on a specific surface. The surface excites plasmons, ultra-short electrowaves, that can transfer energy (or information) much more efficient than for example light. The same way that specific fluoresent probes emit specific lightwaves, the specific chromophores make specific Raman-shifts. Furthermore, higher peaks means that initially more probes were captured. This can be viewed in the figure I used from the article of Christopher D. Syme et al., in Analytical Chemistry 2012.

Screen Shot 2013-01-25 at 16.23.29

 

Next questions are, what is the specificity and sensitivity of SERRS? Is it practical? Can SERRS quantify PCR products? How much does it cost? In part 2 of this blog, I will go deeper into these questions.

To bring this blog to a conclusion, what are so far the advantages and disadvantages of SERRS? The main disadvantage is that SERRS is an end-point analysis system. A molecular biologist likes to observe the amplification curves of a real-time PCR. The shape of the curves contains much information on the performance of the reaction. This, unfortunately, is not the case with the PCR-SERRS method.  The main advantage is the increased discriminatory power in multiplex assays: you are not limited anymore to the light spectrum.

More (dis)advantages in the next part of the PCR-SERRS blog.

Emerging Nucleic Acid-Based Tests for Point-of-Care Detection in low-resource setting.

The Am. J. Trop. Med. Hyg. published a Review entitled: Emerging Nucleic Acid-Based Tests for Point-of-Care Detection of Malaria.
According to the review, Point-of-Care (POC) amplification techniques such as NASBA and, in particular LAMP, will have important benefits for their potential utility in low-resource settings. Real-time PCR based techniques, however, have some major drawbacks to be used at POC: it requires sophisticated temperature and fluorescence detection systems.

Recently I had a conversation with a ‘biohacker’. Together with two others they build at home a PCR device (Amplino) including LED and detector, costing around €50,-. Their vision is to deliver affordable, easy to use malaria diagnostic POC test to the developing world.

Such closed tube PCR devices could work out for malaria, but perhaps also for leishmaniasis, sleeping sickness or even viral hemorrhagic fevers. Even more, by making use of the sensitive detection levels, it might be sufficient to collect specimens by non-invasive methods. For example, instead of performing lumbal punctures or drawing blood, the diagnosis might also be performed on bodily secretions.

How would a POC PCR-device be used in practice? The device will need a power source and an interface to validate the collected data. Most of the power is used for temperature control. Rechargeable batteries are not recommended because they are expensive and they might be depleted before the analysis is complete. A more sustainable power source would be from car-batteries. Either the POC-PCR device is directly connected to the battery under the hood or to a separate battery that can be recharged for example by using a solar panel.

Point of Care PCR

anaconda_v2An interface interprets the raw data (the detected excited light from a probe) and presents all the necessary information to the user. For validating the results, the interface needs to present sample identification, a graphical view of the amplification data, a Cycle treshold value and overview of the controls. The interface can be installed as a software package on a laptop. The laptop is directly connected to the PCR-device. A disadvantage is that all the patient- and analysis data is stored on the laptop, which is prone to theft and defective hard disks. An alternative is to send the raw data directly from the PCR-device by the mobile or satellite network to an internet-server. Using a web-based application, the data can be accessed via a web-browser on your mobile phone or tablet. The data is securely stored and accessible by multiple users.

Slowly the advances in molecular diagnostics are moving towards the rapid developments in information technology. Launching a PCR device that can be used in the middle of nowhere, that would be an invention that would leave all other POC tests miles behind. It would be not only a profit for medical diagnostics. The POC PCR devices can be applied for example phytopathology, water-treatment and veterinary outbreak-control.

Part 1    Part 2

home-made PCR devices

Three young developers build a PCR device costing around €50,-. The device, named Amplino, is made from easy to obtain components. A hairdryer is duck taped to the sample holder. Cookie crumbles over the circuit board. Somewhere an eppendorf lights up.

The concept of a cheap and easy to build PCR device proved to work. Around the world Gyro Gearlooses independently build homemade PCR devices. For example. the OpenPCR machine, both hardware and software, is open source and can be assembled from parts by yourself. A kit with all the necessary parts can be ordered for $500,-. However, unlike Amplino, it doesn’t detect light emission, crucial for clean closed tube DNA detection.

Another limitation of pocket-PCR devices is that (relatively) clean isolated DNA is needed for processing. Although DNA isolation can also be achieved using chemicals from your kitchen cabinet, the tricky part is to get the isolated DNA clear from inhibiting compounds and free from contamination.

Ideally, samples are added to a cassette, containing all the necessary components. Put it in a machine and get the results within two hours. GeneXpert system from Cepheid comes with a nifty machine that performs molecular analysis straight on patient material. However, the kits are quite expensive, even for laboratories in industrialized countries.

Providing one hundred dollar PCR analyzers to poor and endemic countries is beneficial for the public health and it can give a boost to a new market segment. If the PCR-project would work, it will encourage other entrepreneurs to think out of the box and come up with smart alternatives for expensive or failing diagnostic tools.

Part 1   Part 2