WO2006082496A2

WO2006082496A2 - Method for determining hiv-1 viral load

Info

Publication number: WO2006082496A2
Application number: PCT/IB2006/000182
Authority: WO
Inventors: Natela Rekhviashvili; Gwynneth Stevens; Wendy Stevens
Original assignee: National Health Laboratory Service; University Of The Witwatersrand
Priority date: 2005-02-01
Filing date: 2006-02-01
Publication date: 2006-08-10
Also published as: WO2006082496A8; AP2007004111A0; ZA200706396B; WO2006082496A3

Abstract

The invention provides an HIV-1 viral load assay using real-time reverse transcriptase polymerase chain reaction (RT PCR). The assay includes the steps of extracting viral RNA from a plasma sample from a patient; adding at least two primers which are specific to a highly conserved region of the HIV-1 gag gene to the viral RNA from the patient, at least one of the primers being fluorescently-labeled; amplifying the viral RNA from the patient using a real-time reverse transcription (RT) polymerase chain reaction (PCR) method; and applying the results from the amplification step to a standard curve prepared using at least one external standard of known concentration, the external standard comprising an external standard molecule including a RNA sequence equivalent to a highly conserved region of the HIV-1 gag gene. The fluorescently-labeled primer is typically a LUX primer. An internal control may be added to a duplicate sample of viral RNA from the patient, so as to detect false negative results. A kit for performing the assay is also described.

Description

METHOD FOR DETERMINING HIV-1 VIRAL LOAD

BACKGROUND OF THE INVENTION

The invention provides an assay for determining HIV-1 viral load in a patient.

Plans for a national roll-out of a nti retroviral (ARV) therapy for HIV-1 positive patients have been initiated and/or implemented in South Africa and many other countries. Implementation of effective ARV therapy in resource limited settings requires both laboratory infrastructure and the development of cost-effective techniques for diagnosis and monitoring of patients on therapy to support effective clinical management of infected individuals. In particular, determination of HIV-1 burden in plasma is valuable for assessing the efficiency of ARV therapy and disease progression. The "gold standard" HIV-1 viral load tests are well validated and FDA approved, but unaffordable for the public sector in developing countries. Much effort locally and internationally is being expended to develop in-house affordable HIV-1 viral load tests.

SUMMARY OF THE INVENTION

According to a first embodiment of the invention, there is provided an assay for determining HIV-1 viral load in a patient, the assay including the steps of: a) extracting viral RNA from a plasma sample from the patient; b) adding at least two primers which are specific to a highly conserved region of the HIV-1 gag gene to the viral RNA from the patient, at least one of the primers being fluorescently-labeled; c) amplifying the viral RNA from the patient using a real-time reverse transcription (RT) polymerase chain reaction (PCR) method; and d) applying the results from the amplification step to a standard curve prepared using at least one external standard of known concentration, the external standard comprising an external standard molecule including a RNA sequence equivalent to a highly conserved region of the HIV-1 gag gene.

The real-time reverse transcription (RT) polymerase chain reaction (PCR) method may be performed on a real-time instrument, such as a LightCycler version 2 from Roche, or any other real-time PCR instrument that is compatible with a LUX primer detection format.

A real-time standard curve may be produced during the performance of the assay, or alternatively, a pre-prepared standard curve may be used. The pre-prepared standard curve may be supplied in electronic format. Typically, a number of standard pre-prepared curves will be provided to allow for variations in PCR efficiency.

The fluorescently-labeled primer used in amplifying the viral RNA from the patient may be a LUX primer. The primers may form a primer set having the following nucleotide sequences: 5' - ACA TCA AGC AGC CAT GCA AAT -3" (SEQ ID NO: 1 )

5¹- GAG AGA ATG TCA CTT CCC CTT GGT TCT CT(FAM) C -3' (SEQ ID NO: 2)

The sequence of the external standard RNA molecule is typically as follows: 5' - ACA UCA AGC AGC CAU GCA AAU GUU AAA AGA UAC AAU CAA UGA GGA GGC UGC AGA AUG GGA UAG AAU ACA UCC AGU ACA UGC GGG GCC UAU UGC ACC AGG CCA AAU GAG AGA ACC AAG GGG AAG UGA CA - 3' (SEQ ID NO:15)

The external standard may be amplified using primers having nucleotide sequences of SEQ ID NOS: 1 and 2.

An internal control may be added to a duplicate sample of viral RNA from the patient.

The internal control may be a synthetic DNA or RNA molecule having a scrambled

DNA or RNA sequence which comprises the same base pair composition as SEQ ID NOS: 5 or 17, but with a different sequence. For example, the nucleotide sequence of the internal control may correspond to the sequence of SEQ ID NO: 14, or an RNA equivalent thereof (SEQ ID NO: 16). The sequence of the internal control RNA molecule may be as follows:

5' -AAC CUA UCC GGA CAA UAA CGA GUA GGA CAU GAC GAG AAU ACA UGA UAA GUA UGU GAG AGG AGC AUU CAU ACU CGA GAC AGC UCA CGU AGA GCA UCG UCG CAA GCG CAA GCG AGA AAA AGA UCG GAC GGG AG - 3' (SEQ ID NO: 16)

The internal control is typically a double stranded DNA molecule.

The internal control may be used to control for false negative results, typically due to PCR inhibition when the viral load in the patient sample is too high. Thus, if the duplicate with the internal control shows no amplification, then the patient's sample will be diluted and the assay will be repeated using the diluted sample.

Different primers to those used to amplify the viral RNA may be used to amplify the internal control. The fluorescently-labelled primer for use in amplifying the internal control may be a LUX primer. The primer may have a nucleotide sequence selected from the group consisting of: 5' - AAC CTA TCC GGA CAA TAA CGA - 3' (SEQ ID NO: 3); and

5'- AGC GAG ACT CCC GTC CGA TCT TTTTCT(JOE) CGC T-3' (SEQ ID NO: 4).

According to a second embodiment of the invention, there is provided a kit for performing the assay described above, the kit including one or more of the following: instructions for performing the method, optionally in computer readable format; one or more standard curves in an electronic format; a stock solution of one or more external standards (calibrators) of known concentration that were used to prepare the standard curve(s), or for preparing a standard curve; an internal control as described above; a PCR master-mix; and/or one or more sets of primers or sequences thereof, such as SKT145/SKT150 (FAM) and SK IPC145/SK IPC150 (JOE). BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows the distribution of log (Amplicor), log (invention) ("log (LUX)") and log (invention) ("log (LUX)") with upper limit set at 750000 copies/ml; Figure 2 shows scatter plots for log Amplicor verse log Invention and a fitted line, all range values (A), and assay of the invention with the upper limit at

750 000 copies/ml (B);

Figure 3 shows a Bland-Altman plot for the difference Log (Roche)-Log

(Invention) by Log (Roche) the standard; Figure 4 shows percentage similarity histograms and scatter plots for full range, and invention assay upper limit 750 000;

Figure 5 shows Box-and-Whisker Plots for Log (Roche) and Log (LUX) for full range values (A,C) as well as for ranges 400-750 000 copies/ml for both assays (B, D), by visit; and Figure 6 shows a quantitative analysis for RT-PCR samples using the assay of the invention.

DETAILED DESCRIPTION OF THE INVENTION

An HIV-1 viral load assay using real-time reverse transcriptase polymerase chain reaction (RT PCR) is described herein.

An HIV-1 viral load test assay (generally referred to below as the "invention assay" or the "assay of the invention") is typically performed on the LightCycler version2 platform (Roche Applied Science, Mannheim, Germany). The assay uses a generic instrument and generic quantitation software, which makes it easily transferable to different laboratories. Thus, it is not necessary to purchase a new and specific instrument and software to perform the assay, as is required for existing assays, such as the Roche Amplicor HIV-1 Monitor v1.5 test, which requires a COBAS PCR instrument in order to perform the assay.

Patient samples are tested in duplicates - with and without an internal control. The samples that have no internal control are used for determination of the copy number of virus (i.e. for quantitation). The duplicate samples containing the internal control are used solely as a control for PCR inhibition and to prevent false negative results. A concentrated stock solution of external standard RNA may be supplied. To perform real-time quantitation of viral copies, two alternatives are possible - to run about six standards together with the samples (i.e. so as to obtain a real-time standard curve), or to run only one or two standards that will allow the user to import a standard curve (prepared previously in a different experiment) into the run after it has been completed.

For example, when using a pre-prepared standard curve, after the RT-PCR run has been completed, the pre-prepared standard curve is imported for quantitative analysis. In order to accomplish this, each new RT-PCR run must contain at least one standard (calibrator) of any concentration that had been used to generate one or more standard curves. It is advisable to include two calibrators in the run in case one of them fails to amplify. If the standard fits one of the imported standard curves, then this curve is the one which is selected for quantitation (Figure 6).

Since the external standard for quantitative real-time RT-PCR represents a synthetic RNA molecule, it would be conventional to use the same RNA as a standard in combination with an imported standard curve. In this assay, an internal control represents dsDNA (double stranded), although it is conventional to use an RNA internal control for RT-PCR. The reasons for replacing the RNA internal control with DNA are mainly to reduce the cost of the test, to make the production of the internal control stocks easier and more affordable, and to make daily routine handling and storage of the internal control easier, as DNA is more stable than RNA and can be stored at -2O⁰C (instead of -7O⁰C for RNA). The use of DNA as an internal control is unconventional and to the inventor's knowledge has not been reported previously.

The function of the internal control in the assay of the invention is to prevent false negative results due to PCR inhibition. Since viral RNA is extracted using column- based technology that provides highly purified templates, the main reason for PCR inhibition could be a very high viral load. When RT-PCR is overloaded with RNA template, it inhibits both the RT (reverse transcription) step and the subsequent cDNA amplification step. Therefore, the DNA internal control will fail to amplify if the RT-PCR is inhibited due to very high viral load. Samples with no amplification of both the wild type and internal control will be interpreted as false negative and will thus be repeated after diluting the template. Samples will be reported as negative if only the internal control is amplified.

The external standard was cloned and an in vitro RNA was produced specifically for the assay. The internal control was also designed, cloned and produced specifically for the assay. In other words, both clones are unique for this viral load assay.

Commercial HIV-1 viral load tests that are based on NAT (Nucleic acid Amplification Techniques, i.e. RT-PCR, Nucleic Acid Sequence Based Amplification (NASBA)) and so forth and real-time detection of amplification product use primers for amplification and different types of probes for detection. In the HIV-1 viral load assay of the invention, LUX primers are used for the fluorescently-labeled primer, which combine the functions of primers and probes.

Light upon extension (LUX) primers represent a recent advance in real-time detection technology. This detection format includes one single-labelled, self-quenched primer and an unlabelled counterpart. Changes in emission of fluorescence are affected by the primary and secondary structure of a LUX primer. The labelled 3'-end of a LUX primer requires guanosine bases near the conjugated fluorescent tag. The 5'-end of this fluorogenic primer is modified by the addition of a short sequence that is complementary to the labelled 3'-end of the primer. This 5' tail allows a LUX primer to assume a hairpin configuration, which causes a fluorescence quenching effect. During PCR, when a fluorogenic LUX primer binds to the complementary sequence and becomes linear, the fluorophore is de-quenched and the levels of fluorescence increase by 10-fold (Nazarenko I, et a/., 2002 (1); Nazarenko I, et a/., 2002 (2); Lowe B, et al., 2003; Chen R, et al., 2004; Sharkey F, et al., 2004; Donia D, et al., 2005). LUX primers represent one of the most affordable options amongst product-specific real-time detection formats, since they do not require an additional quencher or a detection probe. Apart from substantially reducing the cost of the test, LUX primers make it quicker and easier to assemble the reaction mix and make the assay more suitable for detection of the different subtypes of HIV-1.

This assay is more affordable then any of the current gold standard tests. For example, preliminary cost analysis of the reagents showed that the cost of the viral load assay of the invention is about half of the cost of the Roche Amplicor test. It is envisaged that a kit will be provided for carrying out the assay described above. The kit will initially include instructions for performing the assay (which may be in computer readable form to allow the viral copy number to be determined by a computer); one or more pre-generated standard curves, preferably also in computer readable format (although it will also be possible for users of the assay to prepare and save their own standard curves from the purchased external standard RNA); one or more external standards of known concentration that were used to prepare the standard curve(s),or for use in preparing a standard curve; an internal control as described above; a PCR master-mix of the type that is commercially available; and/or two sets of primers or the sequences of these primers, such as: SKT145/SKT150 (FAM) and SKIPC145/SKIPC150 (JOE). One or more of the above items may also be sold individually, such as the external standards, primer sets, PCR master-mix or internal controls.

Methodology and assay design

RNA Extraction

Potentially, any column or magnetic bead-based kit for manual or automated extraction of HIV-1 RNA may be used for a front end of the assay of the invention. The main requirement for the extraction procedure is an isolation of a highly pure and concentrated nucleic acid. For the development and evaluation of the viral load assay, two types of extractions were used, as described in more detail below.

Viral RNA was extracted using 500 μl of plasma from VQA (Virology Quality Assurance Laboratory, Rush-Presbyterian-St Luke's Medical Centre, Chicago, Illinois), HIV-1 RNA copy control samples and a QIAamp^® Viral RNA mini kit (Qiagen GmbH, Hilden, Germany). Manual extraction was performed according to the manufacturer's (Qiagen) instructions. Prior to extraction, plasma was centrifuged for 60 minutes at 23 000 x g in a bench top centrifuge pre-cooled to 4⁰C. HIV-1 RNA from randomly selected patients' samples was also extracted manually using the QIAamp kit (Qiagen). HIV-1 RNA from longitudinally followed-up patients' samples was extracted from 200 μl of plasma (without prior centrifugation) using a MagNA Pure LC Total Nucleic Acid Isolation kit and a MagNA Pure LC Instrument (Roche Applied Science, Mannheim, Germany) for automated extraction. The procedure was performed according to the manufacturer's instructions (Roche). Extracted RNA was kept on ice if used immediately as a template for RT-PCR, or stored at -7O⁰C for later use.

Real-time Quantitative RT-PCR (Reverse Transcriptase Polymerase Chain Reaction)

A Human Immunodeficiency Virus-1 (HIV-1 ) PCR product, i.e. so-called wild type product, was amplified using a set of primers SKT145 and SKT150. Forward primer SKT145 represents a truncated version of primer SK145. The second, reverse primer SKT150 represents a modification of the reverse primer SKCC1 B used in the Roche Amplicor assay. Primer SKT150 is positioned more internally to the primer SKCC1 B and partially overlaps with it. Further details on this primer set are described in Rekhviashvili N, et a/., 2006, the entire contents of which are expressly incorporated herein by reference. These primers are specific for the highly conserved region (i.e. a region with a minimal polymorphism (diversity) of viral RNA sequence) of the HIV-1 gag gene. This allows amplification of PCR product from different subtypes of HIV-1. Modifications were made to the primers in order to convert them to a LUX (Light upon Extension) format (Nazarenko I, et a/., 2002 (1 ), Nazarenko I, et al., 2002 (2)). According to the manufacturer's recommendations (Invitrogen), the reverse primer SKT150 was modified by the addition of the 7 nucleotide (7 bp) tail to the 5' end and a fluorescent dye FAM (Invitrogen Corporation, Carlsbad CA, USA) to its 3' end.

A different set of LUX primers, SKIPC145 and SKIPC150, was designed for amplification of an internal control (IC). The reverse SKIPC150 primer has a 7 nucleotide tail at its 5' end and a fluorescent tag JOE (Invitrogen Corporation, Carlsbad CA, USA) at the 3' end.

Wild type RT-PCR products amplified with LUX primers SKT145 and SKT150 (FAM) were viewed in channel 530nm of the LightCycler version 2 (Roche Applied Science,

Mannheim, Germany) instrument and amplicons of the internal control generated with SKIPC145 and SKIPC150 (JOE) were viewed in channel 560nm. Different fluorescent dyes (i.e. FAM and JOE) attached to the LUX primers allowed detection and differentiation of the wild type/external standard and the internal control amplification products using different optical channels of the instrument. LUX Primers:

SKT145: 5' - ACA TCA AGC AGC CAT GCA AAT -3' (SEQ ID NO: 1 ) SKT150: 5'- GAG AGA ATG TCA CTT CCC CTT GGT TCT CT(FAM)C -3'

(SEQ ID NO: 2)

SK IPC145: 5' - AAC CTA TCC GGA CAA TAA CGA - 3' (SEQ ID NO: 3) SK IPC150 5'- AGC GAG ACT CCC GTC CGA TCT TTT TCT(JOE) CGC T-3" (SEQ ID NO: 4)

Real-time quantitative RT-PCR was performed in a one step assay format. A total reaction volume of 50 μl contained 28 μl of a master mix and 22 μl of extracted viral RNA template. The RT-PCR master mix consisted of the LUX primers, heat-labile uracil-DNA glycosylase (HL UNG) and RT-PCR reaction mix (including enzymes, dNTPs, MgCI₂, and other components) provided in a QuantiTect™ Probe RT-PCR kit (Qiagen GmbH, Hilden, Germany). In duplicate reactions containing 2 μl of internal control cDNA, the volume of RNA template was 20 μl. The RT-PCR reaction mix was assembled according to the manufacturer's instructions (Qiagen). Primers SKT145/150 were used at 0.5 μM and primers SK IC145/150 were used at 0.5 μM concentration. To prevent RT-PCR carry-over contamination with PCR product amplified in a previous run, a heat-labile uracil-DNA glycosylase (HL UNG) (Roche Applied Science, Mannheim, Germany) was added to the master mix at a concentration of 1 U per reaction. HL UNG enzymatically digests any PCR product that could possibly contaminate the new master mix.

Thermal cycling profile on the LightCycler version 2 (Roche Applied Science, Mannheim, Germany) platform comprised an RT step at 55⁰C for 50 minutes, initial activation at 95⁰C for 15 min, 10 cycles of 95⁰C for 5 s, 52⁰C for 15 s, 72⁰C for 20 s, 78⁰C for 3 s, followed by 50 cycles of 95⁰C for 5 s, 55⁰C for 15 s, 72⁰C for 20 s and 78⁰C for 3 s. Ramping rates were 20°C/s for all the steps of the PCR cycle. Fluorescent signals were acquired during the additional heating step (78⁰C for 3 s). A melting curve cycle, for product identification, included 95⁰C for 0 s, 5O⁰C for 1 min and heating up to 9O⁰C with continuous fluorescence measurement. External Standard (ES) for Real-time Quantitation

The external standard represents an in vitro transcribed RNA molecule, which includes a sequence identical to a 128 bp region (SEQ ID NO: 15) of the HIV-1 gag gene, flanked by primers SKT145/150. As mentioned above, this is one of the highly conserved regions of the viral genome that will allow amplification and detection of PCR product across different HIV-1 subtypes and polymorphisms within each subtype.

The sequence of the RNA external standard is as follows:

5¹ - ACA UCA AGC AGC CAU GCA AAU GUU AAA AGA UAC AAU CAA UGA GGA GGC UGC AGA AUG GGA UAG AAU ACA UCC AGU ACA UGC GGG GCC UAU UGC ACC AGG CCA AAU GAG AGA ACC AAG GGG AAG UGA CA - 3' (SEQ ID NO:15)

Alternatively, a DNA external standard molecule with the following sequence could be used:

3'- ACA TCA AGC AGC CAT GCA AAT GTT AAA AGA TAG AAT CAA TGA GGA GGC TGC AGA ATG GGA TAG AAT ACA TCC AGT ACA TGC GGG GCC TAT TGC ACC AGG CCA AAT GAG AGA ACC AAG GGG AAG TGA CA- 5" (SEQ ID NO: 5)

The amplification product (amplicon) of HIV-1 generated with primers SKT145/150 was verified by conventional gel electrophoresis and purified from a 1 % agarose gel using a MinElute™ Gel Extraction kit (Qiagen, GmbH Hilden, Germany). The amplicon was then cloned into a p-GEM T-Easy vector (Promega, USA). A few random clones were sequenced in order to confirm the integrity of the insert. Sequencing was performed on an ABI PRISM^® 3100 Genetic Analyzer (Foster City CA, USA) using a Big Dye^® Terminator version 3.1 cycle sequencing kit from Applied Biosystems (Foster City CA, USA). Commercially available primers specific for the T7 and SP6 promoter regions of the p-GEM vector were used as sequencing primers. After the sequence of the clones was confirmed, one of these clones - p- GEM-HIV (A5) was selected for future applications. A linearized plasmid of p-GEM- HIV (A5) clone served as a template for the external standard RNA in vitro transcription using a RT-PCR Competitor construction kit (Ambion, Austin Texas, USA) according to the manufacturer's instructions. Newly synthesized external standard RNA was purified using a RNeasy^® Mini kit (Qiagen, GmbH Hilden, Germany). The yield of the external standard RNA transcript was estimated using a RiboGreen quantitation kit (Molecular Probes, USA) and concentration of the stock solution was calculated in copies/μl.

Serial dilutions of external standard RNA covering a range from 4x10⁶ to 4x10² copies per ml were used to generate a number of standard curves for quantitative real-time RT-PCR (Rekhviashvili N et a/., 2006 (3)). Each of the six standards was spiked with 8 ng of MS2 phage RNA (Roche Applied Science, Mannheim, Germany) as stabilizing background RNA.

Figure 6 shows a standard curve prepared as described above and imported into the RT_PCR run. ES1-3 represent RNA standards, and in this case ES1 was used to import the standard curve into the run. The negative control and the blank sample showed no amplification product. Copies of HIV-1 RNA were quantitated in the samples in copies/μl and then converted into copies/ml.

Internal Control (IC)

A non-competitive internal control (IC) (i.e. amplified using a different set of primers to the external standard and viral RNA) representing a synthetic 128 bp DNA molecule with an artificial sequence was designed, having exactly the same base composition and length as the wild type product, but with a different sequence of nucleotides.

Thus, the internal control includes the following nucleotides, but in a different order: DNA: 3'- ACA TCA AGC AGC CAT GCA AAT GTT AAA AGA TAG AAT CAA TGA GGA GGC TGC AGA ATG GGA TAG AAT ACA TCC AGT ACA TGC GGG GCC TAT TGC ACC AGG CCA AAT GAG AGA ACC AAG GGG AAG TGA CA- 5' (SEQ ID NO: 5); or

RNA: 3'- ACA UCA AGC AGC CAU GCA AAU GUU AAA AGA UAG AAU CAA UGA GGA GGC UGC AGA AUG GGA UAG AAU ACA UCC AGU ACA UGC GGG GCC UAU UGC ACC AGG CCA AAU GAG AGA ACC AAG GGG AAG UGA CA- 5' (SEQ ID NO: 17)

An artificial sequence of the internal control was assembled from four oligonucleotides (HIVic S1 , HIVic S2, HIVic A1 and HIVic A2) and served as a template for PCR. The sequences of these oligonucleotides were designed by the inventor and sent to MWG-Biotech AG (Germany) for manufacture. The sequences of the oligonucleotides are as follows:

HIVic S1: 5' - CTA GAA CAT CAA GCA GCC ATG CAA ATG TAG GAC ATG ACG

AGA ATA CAT GAT AAG TAT GTG AGA GGA- 3' (SEQ ID NO: 6) HIVic A1: p 5' - GAA TGC TCC TCT CAC ATA CTT ATC ATG TAT TCT CGT CAT

GTC CTA CAT TTG CAT GGC TGC TTG ATG TT- 3' (SEQ ID NO: 7) HIVic S2: p 5' - GCA TTC ATA CTC GAG ACA GCT CAC GTA GAG CAG GCT

CGT CGC ACG AGA GAA CCAAGG GGAAGT GAC AC - 3'(SEQ ID

NO: 8) HIVicA2: 5' - TCG AGT GTC ACT TCC CCT TGG TTC TCT CGT GCG ACG AGC

CTG CTC TAC GTG AGC TGT CTC GAG TAT- 3' (SEQ ID NO: 9)

The 5' ends of the oligonucleotides A1 and S2 were phosphorylated in order to facilitate ligation. Oligonucleotides S1 and S2 were initially annealed with oligonucleotides A1 and A2, respectively, to form two double stranded oligonucleotides - S1A1 and S2A2. The double stranded oligonucleotides (i.e. S1A1 and S2A2) containing phosphorylated 5' ends were then ligated with each other to produce a double stranded DNA molecule of the internal control, which was subsequently used as a template for amplification by PCR. An amplification product of a competitive internal control generated using primers SKT145/150 was then cloned into a p-GEM T-Easy vector (Promega, USA). Sequencing of plasmids and gel purification of a linearised plasmid p-GEM-HIV (M1 ) was performed as described above for the external standard. A plasmid containing the sequence of the internal control was used to generate a PCR product of a competitive internal control (i.e. amplified using the same set of primers SKT145 and SKT150 as the external standard and viral RNA). Purified PCR product of the competitive internal control was modified by PCR engineering to produce a non-competitive internal control. The sequences of primers SKT145 and SKT150 were scrambled and these new sequences replaced primer binding sites of the competitive internal control. PCR engineering was performed in two experiments using two sets of primers, respectively: SK IPC 145(A): 5'- AAC CTA TCC GGA CAA TAA CGA GTA GGA CAT GAC GAG

AAT-3' (SEQ ID NOMO) SK IPC 150(A): 5'-C TCC CGT CCG ATC TTT TTC TCG CTG CGA CGA GCC

TGC TCT ACG-3' (SEQ ID NO: 11 ) SK IPC145: 5'- AAC CTA TCC GGA CAA TAA CGA-3' (SEQ ID NO: 12)

SK IPC150: 5'- CTC CCG TCC GAT CTT TTT CTC GCT - 3' (SEQ ID NO: 13)

The first set of primers (A) contains a sequence that is internal to the primers SKT145 and SKT150 and overhangs that represent new, scrambled primer sequences. The second set of primers corresponds to the new, scrambled sequence and was used in a subsequent experiment to select and preferentially amplify the desired product, i.e. a non-competitive IC.

The sequence of the non-competitive internal control is as follows: 5'- AAC CTA TCC GGA CAA TAA CGA GTA GGA CAT GAC GAG AAT ACA TGA TAA GTA TGT GAG AGG AGC ATT CAT ACT CGA GAC AGC TCA CGT AGA GCA GGC TCG TCG CAA GCG AGA AAA AGA TCG GAC GGG AG - 3' (SEQ ID NO: 14)

It is, however, envisaged that the internal control could also be an RNA sequence, such as:

5' - AAC CUA UCC GGA CAA UAA CGA GUA GGA CAU GAC GAG AAU ACA UGA UAA GUA UGU GAG AGG AGC AUU CAU ACU CGA GAC AGC UCA CGU AGA GCA GGC UCG UCG CAA GCG AGA AAA AGA UCG GAC GGG AG - 3' (SEQ ID NO: 16)

The non-competitive internal control was quantified using a PicoGreen DNA quantitation kit (Molecular Probes, USA) and the concentration of the stock in copy per μl was established. The non-competitive internal control was added to the duplicate reaction of each sample 2μl of 10⁶-10⁵ copies/μl working solution per reaction.

Real-Time Quantitation and Detection

Real-time quantitation was performed using software version 4 of the LightCycler v2 instrument (Roche). The external standard RNA was serially diluted (1 :10 and 1 :5) and six standards were used at the following concentrations: 4x10⁶, 4x10⁵, 4x10⁴, 4x10³, 8x10² and 4x10² copy/ml.

Since the concentration of the external standard stock was established in copies/μl (as described above), conversion to copies/ml was done taking into account external standard RNA input volume. This type of calculation was only used at the initial stage of assay evaluation, before patients samples were used. Standard curves generated during the replicate testing of the external standard RNA (Rekhviashvili N₁ et a/., 2006 (3)) were saved and exported for future applications. A number of standard curves were exported, each in two versions - analyzed with "auto" and "fit point" methods, respectively. These standard curves were randomly selected and varied slightly in PCR efficiency. Since RT-PCR runs for HIV-1 viral load may differ in PCR efficiency, which affects quantitation, it is ideal to have a choice of standard curves that could accommodate such variations. Most of the exported standard curves were also re-saved with the corresponding values in copies/μl for determining viral loads in clinical samples (Figure 6). Thus, the viral load (VL) values in samples were detected in copies/μl and then converted to copies/ml using the following formula:

VL (copies/ml)= VL (copies/μl) x elution volume (μl)/ sample volume (ml)

The external standard and the wild type PCR products were viewed and analyzed in the 530nm channel of the LightCycler v2 (LC v2). The internal control was analyzed using the 560 nm channel.

Evaluation of the assay performance - preliminary results

1.1. Preliminary evaluation using external standard RNA

Preliminary evaluation of the assay performance was conducted using synthetic RNA of an external standard (ES RNA). This feasibility study aimed to establish analytical sensitivity, dynamic range and reproducibility of the LUX assay. Rekhviashvili N, et ai, 2006(3) provides detailed information on the feasibility study. In brief, the assay revealed a wide linear (R² = 0.99) dynamic range from 4x10⁶ to 4x10² copies/ml of external standard RNA. The limit of quantitation was set at 400 copies/ml of external standard RNA. Overall variability of the assay (intra- and inter-assay variability) was less than 0.5logio copies of external standard RNA (i.e. no clinically significant variability was observed).

1.2. Evaluation of the LUX assay using patient (or clinical) specimens

1.2.1. Specificity of the invention assay:

A cohort of 30 healthy blood donors obtained from the South African blood bank service was used to evaluate additionally the specificity of the assay. HIV-1 negative status of these individuals was established using the p24 ELISA test.

Viral RNA for the assay was extracted using the MagNA Pure system (Roche), as described above. All samples were tested in duplicates - with and without an internal control - in order to exclude false negative results. IC product was amplified in all the duplicate samples. However, no HIV-1 product was detected in any of the samples using the assay, i.e. no false positive results were observed for this cohort. Therefore, the assay showed good specificity of HIV-1 detection.

1.2.2. Randomly selected patients/ Comparison with gold standard assay:

A cohort of 140 randomly selected patients was used to study the correlation of the developed assay according to the invention (termed the LUX assay) with the in country "gold standard" assay - Amplicor HIV-1 Monitor version 1.5 (Roche Molecular Diagnostics, Branchburg, NJ, USA).

Table 1 : Summary statistics for Roche Amplicor (copies/ml), Log 10 (Amplicor), Invention (copies/ml) and Log (Invention). N = 139

One individual who had an invention assay reading at 29 000 000 copies/ml was removed from the analysis. Though this was a plausible value, since the Roche assay estimated the viral load for this sample at >750 00 copies/ml, it had such high leverage (large influence) that the decision was made to exclude the sample from the statistical analysis. The lower detection limit for both the Roche assay and the invention assay was 400 copies/ml. For the Roche assay, the upper detection limit was set at 750 000 copies/ml, meaning any reading that was 750 000 copies/ml or more was set to 750 000. The upper detection limit for the invention assay was not restricted, and thus had a wider range of from 400 - 5 640 000. The means and medians for the untransformed assays show that the distributions for these are highly skewed (positive skewness), and thus the data was log transformed to reduce the wide range and to make the data assume a roughly normal distribution. All the analyses that follows thus use log (log 10) transformed data.

In Figure 1 , the first graph (A) shows the distribution of Log(Amplicor), the Roche assay; the second graph (B) shows Log(LUX), the invention assay, with the original range of values; and the last graph (C) shows the distribution of Log(lnvention) setting the upper limit to log (750 000). Note the lumping of values at the lower and upper limits, especially for the first and last graphs. This shows the distribution of patients before ARV therapy, with high viral loads, and after ARV therapy, with reduced viral loads (less than 400 copies/ml).

The frequency graph (Fig 1 ) for this cohort of patients (n=139) tested with the Roche assay reflected a typical distribution of viral load values in the population receiving ARV therapy, with a maximum frequency at the lower (2.6 log₁₀) and the higher (5 logio-6-log₁₀) values. The assay of the invention, with the original quantitation range, showed a maximum distribution at 2.6 log₁₀ and 4 - 5 log₁₀ of viral load values.

Correlation between the assay of the invention and the "gold standard" assay was explored using the Pearson Correlation Coefficient and the non-parametric Spearman Correlation Coefficient. A significant linear relationship was shown to exist between the two assays.

For log transformed viral load values (log Amplicor and log Invention), the Pearson correlation coefficient was 0.85, with a corresponding p-value of <0.0001. This suggests a strong linear relationship between the two log transformed variables. Similar results were observed when the Spearman correlation coefficient was used: r²=0.84 (p-value = <0.0001 ).

Regression analysis was performed to quantitate the linear relationship between log₁₀ Amplicor and log₁₀ Invention values (Figure 2).

From the plots of Figure 2, using the full set of recorded values for the assay of the invention gives a false asymptote at log 6 on the y-axis (log(Amplicor)). This is the result of setting Amplicor to have a set maximum value of 750 000 copies/ml, while allowing the invention assay to go beyond 750 000 copies/ml. If the upper limit of the invention assay is set at 750 000 copies/ml, it is possible to achieve a better fit that no longer has this artificialness (Fig 2(B)).

The relationship between log Amplicor and log Invention for the two lines of Figure 2 with the linear functions, can be represented by the following equations: \og\0(Amplicor) = 0.81 + 0.83 log 10(/ux) , and \og{Amplicor) = 0.60 + 0.87 \og(Lux)

The corresponding R-square values are 73% and 74% for full range of the invention assay viral load log values and for the maximum upper value set at log(750 000), respectively. These equations have all highly statistically significant intercept and gradient coefficient values (all p-values < 0.01 ). The gradients (0.83 and 0.87 for the full range and the restricted range, respectively) for the two graphs are close to 1 , indicting good prediction between the two assays.

Method comparisons for agreement, accuracy and precision were also analyzed for this random cohort of samples. The Bland-Altman analysis (Figure 3) measures absolute differences between the two methods. The log transformed values were used.

The mean difference (middle line) for the full range (A) data is 0.039 95% Cl [-0.063 - 0.141], which shows good accuracy. The limits of agreement defined as the mean plus/minus 1.96 times the standard deviation is [-1.154; 1.232], which shows moderate precision (i.e. a 1.1 log difference is a just clinically acceptable difference). The same conclusions are reached if one restricts the invention assay to have a maximum limit of 750 000 copies/ml (B).

The mean difference in both cases above is less than 0.5 Iog10 (not clinically relevant), showing good accuracy. However, the limits of agreement are -1.2, thus indicating moderate precision.

Figure 4 shows percent similarity histograms (1A, 2A) and scatter plots (1B, 2B). The full range graphs (1A, 1 B) have a percent similarity mean of 100.59%, a standard deviation (SD) of 7.53%, and a coefficient of variation (CV) (SD/mean) of 7.74%. The graphs to the right (2A, 2B), with the same upper range values, have a percent similarity mean of 100.15, a SD of 7.53, and a CV of 7.52.

These results corroborate the results obtained in the Bland-Altman graphs, i.e. high accuracy (percent similarity close to 100%) and moderate precision (histogram range values 80% to 120%, similar to the scatter).

1.2.3. Patients on ARV therapy:

A cohort of 55 patients on ARV therapy, followed up longitudinally (i.e. baseline visit, week 4, 8 and 12 visits), was used. This study aimed to determine the usefulness of the invention assay for monitoring patients on therapy. In other words, the new assay must reflect the same downward trend in viral load values as the gold standard test.

The range of viral load values for the Roche assay was <50 to >750 000 copies/ml, while the range values for the invention assay was <400 to 27 000 000 copies/ml. For the samples on the week 0 (baseline) visit, the standard version of the Roche assay was used (<400 - >750 00 copies/ml) and for the follow-up visits (weeks 4-12), only the ultra sensitive version of the Roche assay was used (<50 - >100 000 copies/ml). This made comparisons difficult. The viral load count data was highly skewed, so all data was log 10 transformed before being analyzed. To make analyses more comparable, any value below 400 copies/ml was set to 400 (lower limit for the invention), while the upper limit for the invention assay was set to 750 000 (i.e. better comparable to the upper limit for the Roche assay). Table 2: Distribution of Roche asay and invention assay (copies/ml), and log (Roche), log (Invention) for both original scale values, and for value range log (400) to log (750000)

Mean SD median Q3-Q1 1^bl-99^ιπ Percentiles

Roche assay Week-0 351617 269060 277000 118000-678000 6320 - 750000

Week-4 2258 13432 254 91 - 569 50-100000

Week-8 165 244 59 50 - 203 49-1580

Week-12 2048 13498 50 50-54 50-100000

Log (Roche) Week-0 5.36 0.48 5.44 5.07 - 5.83 3.80 - 5.87

Week-4 2.45 0.60 2.40 1.96-2.76 1.70-5.00

Week-8 2.00 0.39 1.77 1.70-2.31 1.69-3.20

Week-12 1.90 0.57 1.70 1.70-1.73 1.70-5.00

Log (Roche) Week-0 5.36 0.48 5.44 5.07 - 5.83 3.80 - 5.87

400-750000 Week-4 2.76 0.37 2.60 2.60 - 2.76 2.60 - 5.00 copies/ml Week-8 2.62 0.09 2.60 2.60 - 2.60 2.60- 3.20

Week-12 2.67 0.37 2.60 2.60 - 2.60 2.60- 5.00

Invention assay Week-0 1067657 1944474 271800 97440-1026000 21600-11910000

Week-4 491302 3640625 400 400 - 400 400- 27000000

Week-8 417 124 400 400 - 400 400-1320

Week-12 644 1807 400 400 - 400 400-13800

Log (Invention) Week-0 5.53 0.67 5.43 4.99 - 6.01 4.33 - 7.08

Week-4 2.69 0.65 2.60 2.60 - 2.60 2.60 - 7.43

Week-8 2.61 0.07 2.60 2.60 - 2.60 2.60-3.12

Week-12 2.63 0.21 2.60 2.60 - 2.60 2.60-4.13

Log (Invention) Week-0 5.39 0.47 5.43 4.99 - 5.87 4.33 - 5.87

400-750000 Week-4 2.66 0.44 2.60 2.60 - 2.60 2.60 - 5.87 copies/ml Week-8 2.61 0.07 2.60 2.60 - 2.60 2.60-3.12

Week-12 2.63 0.21 2.60 2.60 - 2.60 2.60-4.14

In Table 2, week 0 represents the baseline viral loads for patients just before commencement of treatment. The median values at baseline for both the Roche assay and invention assay untransformed data are lower than the means, signifying positive skew-ness. Log base 10 transformations were thus performed to normalize the data. Once treatment had commenced, the viral loads dropped significantly in both assays, from median log 5.44 to 2.40 and 5.43 to 2.60 for the original scale Roche and invention assays, respectively. The ultra-sensitive Roche test was used for the patients monitoring from week 4 onwards. Therefore, from weeks 4-12, the viral load values for the invention assay (non-ultra sensitive test with LOG of 400 copies/ml) was compared to the ultra sensitive version of the gold standard test, with the quantitation range of > 100 000 - < 50 copies/ml. This explains more variable viral load values detected by the Roche assay for the weeks 4-12 samples. However, for the invention assay, once treatment started, all values except for one individual read the minimum detected limit of 400 copies/ml for this assay (see Box plots (Figure 5)). Table 3 shows the three individuals with non-400 copies/ml on the invention assay after treatment had commenced.

Table 3: Follow-up samples showing differing viral load values

Observation Roche log(Roche) Lux Luxjog Visit

86 100000 5.00000 13800 4.13988 week 12 162 100000 5.00000 27000000 7.43136 week 4 199 50 1.69897 1320 3.12057 week 8

Individual 199 had a Roche count of 50 copies/ml on the ultra-sensitive Roche assay with a corresponding invention assay value of 1 320 copies/ml, which is probably consistent. It is likely that this patient could have a viral load value of <400 copies/ml, if the standard (non ultra sensitive) Roche test had been performed. This increase in the viral load is not significant enough on its own to change the clinical decision. In this case, the corresponding CD4 counts would be considered and the repeat sample of the patient would be requested. The same patient, at the week 12 visit, revealed undetectable viral load values on both the Roche (ultra) and the invention assays. Another individual (not shown in Table 3) had a viral load of 8 290 on the Roche assay, and 400 copies/ml on the invention assay scale at 12 weeks. This could be explained by the fact that the ultra sensitive test may produce higher values at the lower viral load range. In addition, the same specimen had been tested by the invention assay retrospectively, after storage at -7O⁰C and an additional freeze- though procedure, which could have affected the sensitivity of detection. In conclusion, both of these individuals, irrespective of log difference in viral load values on a particular visit, showed an overall trend of a greatly suppressed viral load in comparison to the baseline reading. Both individuals did not show the spike characteristic shown by the other two individuals.

The other two patients (Table 3) showed spikes (i.e. significant increase) in viral load readings on both assays, and were thus individuals who were most likely not responding to treatment. These results are of a great importance since they show that the invention assay can detect an increased viral load in the same samples as the gold standard, and thus can timeously identify treatment failure in patients.

It is important to note that since there was hardly any variation in the invention assay after starting treatment, it is not possible to use regression techniques (mixed or random effect models) to investigate the association between the Roche and invention assays.

Figure 5 shows Box-and-Whisker plots over time for the log transformed Roche and invention assays. The three individuals also show these spikes on the Roche assay, and thus the two assays picks up similar fluctuations in viral load.

From Figure 5, it can be seen that once treatment starts, viral load rapidly drops below the quantitation limit on both assays. As shown in Table 1 , there is hardly any variability after treatment starts, as the invention assay reaches the 400 lower detection limit at 4 weeks and remains there for all observations except for the individuals mentioned above. The Roche assay similarly shows a dramatic decrease in patients' viral loads in response to therapy starting from the week 4 visit. The spikes in viral load values were detected by both assays in the same samples. Therefore, the invention assay and the gold standard assay show similar trends in monitoring patients on ARV therapy. Summary

• Data analysis on log base 10 transformed values was performed for both Roche and the invention assays. For the invention assay, two sets of analysis have been presented, one with the log original range scale, and the other setting the upper scale limit to log 750 000 copies/ml to make the limits similar to the Roche assay, thus making the comparison better.

• The Pearson correlation coefficient and the Spearman correlation coefficient for the relationship between log (Roche) and log (Invention) are both above 84%, suggesting a strong linear relationship.

• Assuming the original invention assay scale, the relationship between log(Roche) and log(lnvention) is given by: loglO(Amplicor) = 0.81 + 0.83 log lθ(lnvention)

• This linear equation has an R-square value of 73%, meaning we can explain 73 % of the variation in log (Roche) by log (Invention).

• Setting the upper limit of the invention assay to 750 000 copies/ml so that it becomes similar to Roche would give a Pearson Correlation Coefficient of 86%, and a Spearman correlation of 83% with a regression line

\og(Amplicor) - 0.60 + 0.87 \og(Invention) • This line has an associated R-square value of 74%, and thus we can explain 74% of the variation of Log (Roche) by log (Invention).

• The difference plots show good accuracy, mean difference 0.0391 95% Cl [- 0.0630; 0.1412], with moderate precision, limits of agreement [-1.1539; 1.2323].

• Percentage similarity analysis as well as the difference plots analysis, shows good accuracy (mean close to 100), with moderate precision, CV 7.74% for original scales and CV 7.52% for same range scale.

Parallel readings, one by the Roche assay and another by the invention assay, were made at the baseline visit (week 0), at commencement of treatment, and then at weeks 4, 8 and 12. Box-and-Wisker plots for log (Roche) and log (invention), for full range values and for similar range values (400 - 750 000 copies/ml) showed that the only variability occurred at baseline (at the beginning of treatment), and thereafter almost all readings were below 400 copies/ml, the minimum detection limit for the invention, or were clustered around 400 for the Roche assay (ultra-sensitive Roche could read below 400 copies/ml). The assay of the invention can thus be used for monitoring patients' responses to treatment. In a cohort of 55 patients (275 samples) receiving ARV therapy and being followed up longitudinally (5 sequential visits), the invention assay showed a similar downward trend in viral load values over time as the gold standard assay. The two patients that had a significant increase in viral load were detected by both the assay of the invention and the Roche assay.

In conclusion, the assay of the invention combines affordable price, ease of use, and quality that is comparable to the gold standards.

References:

Chen, R., Huang, W., Lin, Z., Zhou, Z., Yu, H., Zhu, D. (2004) Development of a novel real-time RT-PCR assay with LUX primer for the detection of swine transmissible gastroenteritis virus. J. Virol. Meth. 122, 57-61.

Donia, D., Divizia, M., Pana, A. (2005) Use of armoured RNA as a standard to construct a calibration curve for real-time RT-PCR. J. Virol. Meth. 126, 157-163.

Lowe, B., Avila, H.A., Bloom, F. R., Gleeson, M. and Kusser W. (2003) Quantitation of gene expression in neural precursors by RT-PCR using self-quenched, fluorogenic LUX primers. Anal. Biochem. 315, 95-105.

Nazarenko I, Pires R, Lowe B, Obaidy M and Rashtchian A. 2002. Effect of primary and secondary structure of oligodeoxiribonucleotides on the fluorescent properties of conjugated dyes. Nucleic Acid Research 30 (9): 2089-2195 (1 ).

Nazarenko I, Lowe B, Darfler M, lkonomi P, Schuster D and Rashtchian A. 2002. Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore. Nucleic Acid Research 30 (9): e37 (2).

Rekhviashvili N., Stevens G., Scott L., Stevens W, In press. Fluorogenic LUX primer for quantitation of HIV-1 by real-time RT-PCR. Molecular Biotechnology 32:101-109.

Sharkey, F. H., Banat, I. M., Marchant, R. (2004) Detection and Quantitation of Gene Expression in Environmental Bacteriology. Minireview. Appl. Environ. .Microbiol. 70(7), 3795-3806.

Claims

CLAIMS:

1. An assay for determining HIV-1 viral load in a patient, the assay including the steps of:

(a) extracting viral RNA from a plasma sample from the patient;

(b) adding at least two primers which are specific to a highly conserved region of the HIV-1 gag gene to the viral RNA from the patient, at least one of the primers being fluorescently-labeled;

(c) amplifying the viral RNA from the patient using a real-time reverse transcription (RT) polymerase chain reaction (PCR) method; and

(d) applying the results from the amplification step to a standard curve using at least one external standard of known concentration, the external standard comprising an external standard molecule including an RNA sequence equivalent to a highly conserved region of the HIV-1 gag gene.

2. An assay according to claim 1 , wherein the standard curve is a pre-prepared standard curve supplied in electronic format.

3. An assay according to claim 1 , wherein the standard curve is a real-time standard curve prepared during the performance of the assay by the addition of a number of external standards of known concentration to the viral RNA from the sample. i

4. An assay according to any one of claims 1 to 3, wherein the fluorescently- labeled primer is a LUX (light upon extension) primer.

5. An assay according to any one of claims 1 to 4, wherein the primers have the following nucleotide sequences:

5' - ACA TCA AGC AGC CAT GCA AAT -3' (SEQ ID NO: 1 ); and

5'- GAG AGA ATG TCA CTT CCC CTT GGT TCT CT(FAM)C -3' (SEQ ID

NO: 2).

6. An assay according to any one of claims 1 to 5, wherein the external standard RNA molecule includes the following sequence:

5' - ACA UCA AGC AGC CAU GCA AAU GUU AAA AGA UAC AAU CAA UGA GGA GGC UGC AGA AUG GGA UAG AAU ACA UCC AGU ACA UGC GGG GCC UAU UGC ACC AGG CCA AAU GAG AGA ACC AAG GGG AAG UGA CA - 3' (SEQ ID NO:15).

7. An assay according to any one of claims 1 to 6, wherein the external standard is amplified using primers having a nucleotide sequence of SEQ ID NOS: 1 and 2.

8. An assay according to any one of claims 2 to 7, wherein each real-time PCR run contains at least one standard of known concentration that was used to generate the standard curve.

9. An assay according to any one of claims 2 to 8, wherein each real-time PCR run contains at least two standards of known concentration that were used to generate the standard curve.

10. An assay according to any one of claims 1 to 9, wherein an internal control is added to a duplicate sample of viral RNA from the patient and steps (b) to (d) are repeated using the duplicate sample.

11. An assay according to claim 10, wherein the internal control is a synthetic DNA or RNA molecule having a scrambled DNA or RNA sequence which comprises the same base pair composition as SEQ ID NO: 5 or SEQ ID NO: 17, but with a different sequence.

12. An assay according to claim 11 , wherein the nucleotide sequence of the internal control is the sequence of SEQ ID NO: 14.

13. An assay according to claim 11 , wherein the nucleotide sequence of the internal control is the sequence of SEQ ID NO: 16.

14. An assay according to claim 13, wherein the internal control is a double stranded DNA molecule, the sense strand of which has the sequence of SEQ ID NO: 16.

15. An assay according to any one of claims 10 to 14, wherein the internal control is used to detect false negative results.

16. An assay according to claim 15, wherein the false negative results are due to PCR inhibition when the viral load in the patient's sample is too high.

17. An assay according to claim 15 or 16, which includes the steps of:

(e) diluting the extracted viral RNA sample from the patient when the internal control shows no amplification; and

(f) repeating steps (b) to (d) of the assay.

18. An assay according to any one of claims 10 to 17, wherein different primers to the primers used to amplify the viral RNA are used to amplify the internal control.

19. An assay according to claim 18, wherein the primer for use in amplifying the internal control is a fluorescently-labeled primer.

20. An assay according to claim 19, wherein the fluorescently-labeled primer is a LUX (light upon extension) primer.

21. An assay according to any one of claims 18 to 20, wherein the primers used to amplify the internal control include the following nucleotide sequences:

5' - AAC CTA TCC GGA CAA TAA CGA - 3' (SEQ ID NO: 3); and

5'- AGC GAG ACT CCC GTC CGA TCT TTT TCT(JOE) CGC T-3' (SEQ ID

NO: 4).

22. A kit for performing the assay of any one of claims 1 to 21 , the kit including one or more of the following: instructions for performing the assay, optionally in computer readable format; one or more standard curves, optionally in an electronic format; one or more stock solutions containing external standards of known concentration that were used to prepare the standard curve(s), or for use in preparing a standard curve; an internal control; a PCR master-mix; and/or one or more sets of primers or sequences of the primers.

23. A kit according to claim 22, wherein the sets of primers are: SKT145/SKT150 (FAM) (SEQ ID NOS: 1 and 2) and/or SK IPC145/SK IPC150 (JOE) (SEQ ID NOS: 3 and 4).

24. An assay according to claim 1 , substantially as herein described.