The development of simultaneous measurement of viral load and physical status for human papillomavirus 16 and 18 co-infection using multiplex quantitative polymerase chain reaction.

Persistent infection with human papillomavirus (HPV) type 16 and 18 is known to be a major risk factor for cervical cancer. Increased prevalence of co-infection with these high-risk types has been observed in pre-cancerous and cancerous tissues. The determination of physical status and copy numbers of viruses is therefore useful in clinical settings. A simple multiplex quantitative polymerase chain reaction (qPCR) for HPV16/HPV18 co-infection in one tube reaction was established in the present study using TaqMan®-based PCR for E2 and E6 viral DNA. The detection range was up to 106 copies with 100% specificity and high precision (CV of cycle time <0.5%). The analytical accuracy and robustness were verified by competitive assay using an unequal mixture of HPV16/HPV18 DNA. No significant effect was demonstrated when compared with the simplex qPCR. The detection of physical status was evaluated in cervical samples, including 5 pre-cancerous and 15 cancerous samples. No significant difference was observed between simplex and multiplex qPCR (P=0.372). In conclusion, the developed multiplex qPCR method successfully demonstrated the viral status of the common HPV types in one tube. This assay will facilitate viral assessment and monitoring of cervical cancer associated with HPV16 and HPV18 co-infection.

Introduction

Cervical cancer (CXCA) is the fourth leading cause of cancer-associated mortality among females worldwide (1). The majority of CXCA cases are associated with high risk human papillomavirus (HR-HPV) infection. Among HR-HPV, HPV16 and HPV18 account for 70% of all CXCA cases worldwide (1), including in Asian populations (2). Epidemiological data revealed an increase in multiple HPV infection from 10% of CXCA cases in 2005 (3) to 65% in 2016 (3–5). However, these figures may reflect a higher sensitivity of detection, resulting in the co-infection of HPV16 and HPV18 being reported as the most common HR-HPV in Africa (6) and Asia (7), and specifically in Thailand (8), as 31.9, 20.8 and 27.8%, respectively. An increased odds ratio of HPV16 and 18 co-infection compared with single infection was demonstrated by Chaturvedi et al (9) and Trottier et al (10), but the opposite results were reported by Salazar et al (11). These controversial findings implied that other viral parameters, not only HPV genotype, may serve important roles in disease progression.

The physical state of HPV infection occurs as one of two forms: An episomal or an integrated form. The episomal state involves the complete life cycle of viral replication in the infected host cells, whereas the integrated form involves integration of HPV DNA into the host DNA, a major genetic event leading to cervical carcinogenesis (12). Viral integration and viral load have previously been reported to be biomarkers for cancer with high-grade cervical lesions (13). Several methods have been used for the detection of integrated HPV, including polymerase chain reaction (PCR) (14), in situ hybridization (15,16) and amplification of papillomavirus oncogene transcripts (APOT) (17). These are all qualitative measurements. Recently, we established a quantitative PCR of E2 and E6 genes to measure the viral load and physical status of HPV16 DNA in one tube (18). A ratio of E2/E6 gene of 1.0 is used to define the episomal form, while a decreased ratio (less than 1.0) indicates the integrated form, due to deletion of the E2 gene during viral integration. The present study demonstrated the benefit of using viral numbers and physical status as surrogate markers of cancer progression.

To the best of our knowledge, the present study was the first to report the simultaneous measurement of 4 genes, E2 and E6 genes of HPV16 and HPV18, in a single tube. The development of multiplex qPCR in the present study provides a total coverage of 70% of HR-HPV-associated CXCA, including single and co-infection of HPV16 and HPV18. The analytical performance of the multiplex qPCR was evaluated in clinical samples, compared with simplex qPCR.

Materials and methods

Samples

A total of 20 cervical tissues harboring single or co-infection of HPV16 and HPV18 were collected from 5 pre-cancerous lesions (mean, 42.2±6.6 years) and 15 cancerous lesions (mean, 49.5±13.7 years). Samples were collected between 2002 and 2004 under written informed consents approved by the Ethical Committee of Khon Kaen University, Khon Kaen, Thailand (approval no. HE562296) and between 2013 and 2014 approved by the the Ubonratchathani Cancer Hospital, Ubon Ratchathani, Thailand (approval no. CE012/2013). DNA samples were extracted using a QIAamp Viral DNA kit (Qiagen GmbH, Hilden, Germany), according to the manufacturer's protocol. Extracted DNA was used for HPV16 and HPV18 screening by Nested Multiplex PCR, as previously described (19).

Cell culture

The human papillomavirus Caski and HeLa cell lines containing the integrated form of HPV16 (600 DNA copies per cell) and HPV18 (20–50 DNA copies per cell) were used as internal standard for determination of physical status of HPV16 and 18, respectively. HeLa cell line containing HPV18 and CaSki cell line containing HPV16 were used as HPV positive controls. Cells were cultured in 25 cm2 flasks at 37°C with 5% CO2 in Dulbecco's Modified Eagle Medium high glucose (DMEM-HG) media supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (10,000 U/ml penicillin and 10 mg/ml streptomycin; all Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). At 70% confluency, cells were trypsinized with 1 ml of 1X trypsin-EDTA at 37°C for 5 min. The cell pellets were collected and centrifuged at 2,000 rpm for 3 min and subsequently used for DNA extraction.

HPV16 and HPV18 integration status assay

Plasmids containing the whole genomes of HPV16, HPV18, HPV45 and HPV58 (PBR322 for HPV16 and HPV58, PGM4 for HPV18 and HPV45) were provided by Professor Pientong from the Department of Microbiology, Faculty of Medicine, Khon Kaen University. Purified recombinant plasmid copy numbers were estimated by a spectrophotometer concentration measurement (NanoDrop 2000; Thermo Fisher Scientific, Inc.). The DNA calculation formula was 6.02×1023 (copies/mol) xA260 (ng/ml)/(DNA length ×660)=copies/ml. Plasmid DNA was then diluted with sterile water to obtain between 106 and 102 copies, and was used to establish calibration curves for measuring E2 and E6 by multiplex qPCR using a TaqMan® probe assay (Bioneer Corporation, Daejeon, Korea). The oligonucleotide sequences of primers and probes were followed as previously described (20,21), with modifications to the quencher and reporter fluorescent dyes (Table I). E2 and E6 DNA were amplified using a qPCR thermocycler (Exicycler™; Bioneer Corporation). Each 50 µl reaction mixture contained a premix (AccuPower® DualStar™; Bioneer Corporation) with 0.4 µM probes and primers. The PCR reaction was initiated at 95°C for 10 min, followed by 40 cycles at 95°C for 15 sec and 63°C for 60 sec. Reactions lacking DNA template were used as negative controls as previously described. Quantification of E2 and E6 genes were analyzed using the calibration curve plotted between the quantification cycle (Cq) on the x-axis and the logarithm of the standard copy number on the y-axis (102−106 copies). Linear regression equations were estimated as indicated in Fig. 1. The obtained Cq from samples were used to calculate E2 and E6 copies from these equations. Amplification efficiency was determined from the slope of log-linear calibration curve (10−1/slope−1) (22).

Table I.

Oligonucleotide sequences and TaqMan® probes used for amplification of E2 and E6 genes of HPV16 and HPV18.

HPV type	Name	Sequence (5′-3′)	Amplified product length (bp)	(Refs.)
HPV16	HPV16 E2 Forward primer	AACGAAGTATCCTCTCCTGAAATTATTAG	82	Peitsaro et al, 2002 (20)
	HPV16 E2 Reverse primer	CCAAGGCGACGGCTTTG
	HPV16 E2 Probe	(TAMRA)-CACCCCGCCGCGACCCATA-(BHQ)
	HPV16 E6 Forward primer	GAGAACTGCAATGTTTCAGGACG	81
	HPV16 E6 Reverse primer	TGTATAGTTGTTTGCAGCTCTGTGC
	HPV16 E6 gene Probe	(Texus red)-CAGGAGCGACCCAGAAAGTTACCACAGTT-(BHQ)
HPV18	HPV18 E2 Forward primer	AGAAGCAGCATTGTGGACCT	167	Damay et al, 2009 (21)
	HPV18 E2 Reverse primer	GGTCGCTATGTTTTCGCAAT
	HPV18 E2 Probe	(TeT)-TCAACC-CACTTCTCGGTGCAGC-(BHQ)
	HPV18 E6 Forward primer	TCACAACATAGCTGGGCACT	91
	HPV18 E6 Reverse primer	CTTGTGTTTCTCTGCGTCGT
	HPV18 E6 Probe	(FAM)-GCCATTCGTGCTGCAACCGA-(BHQ)

HPV, human papillomavirus; bp, base pairs.

Figure 1.

Standard curves of multiplex quantitative polymerase chain reaction for E2 and E6 genes of HPV16 and HPV18. The analytical range was revealed to be 103 to 106 copies. HPV, human papillomavirus; SD, standard deviation.

Optimization of temperature for multiplex qPCR

Multiplex qPCR temperature for HPV16 (E2 and E6 genes) and HPV18 (E2 and E6 genes) was optimized. A mixture of 105 copies of plasmids containing the whole genomes of HPV16 and HPV18 was used to optimize the annealing temperature from 55 to 64°C. The temperature that produced the lowest Cq was selected as the optimal annealing temperature.

Evaluation of multiplex qPCR performances

Multiplex qPCR analytical range

Ten-fold dilutions of mixed HPV16 and HPV18 whole genomic DNA (from 102 to 106 copies) were used as templates for determining the analytical range.

Multiplex qPCR analytical imprecision

Within-run and between-run precision were each determined using low (103 copies) and high (106 copies) concentrations of whole genome HPV16 and HPV18 DNA mixtures.

Multiplex qPCR analytical specificity

Cross-reactivity with two other HPV genotypes, HPV45 and HPV58, was tested. No fluorescent signal indicated a lack of cross-reaction, and uninfected HPV DNA was used as a negative control.

Competitive effect of HPV16 and HPV18 in multiplex qPCR

A mixture of unequal concentrations of HPV16 and HPV18 DNA (from 103 to 106 copies) was used to evaluate competitive effects in multiplex qPCR. E2 and E6 copies obtained from unequal HPV16 and HPV18 templates (test) were compared to those determined using a single template of HPV16 or HPV18 (control) by a paired t-test. No significant difference (P>0.05) indicated no competitive effect.

Evaluation of physical status using multiplex qPCR in clinical samples

The cut-off value for an episomal form (complete E2 and E6 sequence) was first determined using plasmid DNA containing whole HPV16 and HPV18 genomes. The E2/E6 ratio was calculated by the 95% confidence interval (CI) and used to interpret physical status in clinical samples as previously described (18). To verify the accuracy of the multiplex qPCR of HPV16 and HPV18, 105 copies of Caski and HeLa cells containing pure integrated HPV16 and HPV18, respectively, were prepared according to the DNA calculation formula and used as internal standards for the integrated form. A total of 105 copies of whole plasmid genome HPV16 and HPV18 were used as internal standards for the episomal form. The two cell lines were provided by Professor Pientong. A total of 20 cervical samples with single and mixed infection were used to compare the copy number and physical status between the simplex and multiplex qPCR using paired t-tests and χ2 tests.

Statistical analysis

The linear regression equation was estimated from standard curves between viral copy number (y-axis) and cycle time (x-axis). The percentage of efficiency between 80–120% and correlation (R2) >0.98 were used to determine the standard curve. The comparison of viral copy number between multiplex and simplex was performed using paired t-tests, while physical status was compared by χ2 test. P<0.05 was considered to indicate a statistically significant difference. Statistical analysis was performed using SPSS 19 software (IBM Corp., Armonk, NY, USA) under a Khon Kaen University license.

Results

Multiplex qPCR performance for the detection of HPV16 and HPV18 (E2 and E6 genes)

The optimization of Cq was performed under annealing temperatures from 55 to 64°C, as demonstrated in Fig. 2. The optimal Cq for the E2 and E6 genes of HPV16 was 63°C, whereas the optimal Cq for HPV18 was between 55 and 64°C. Therefore, 63°C was used to construct a standard curve for HPV16 and HPV18 multiplex qPCR, as demonstrated in Fig. 1. According to the guidelines for validation of quantitative PCR methods, linear regression and correlation (R2) analyses for each gene revealed an acceptable efficiency of 109–115% (23). The analytical range was verified at 1,000-1,000,000 copies for HPV16 and HPV18, with average imprecision from 0.42 to 0.50% CV as demonstrated in Fig. 1. The imprecision of measured HPV copies is presented in Table II. The average CV of within-run and between-run was 10.2 and 12.1, respectively. No cross reactivity was observed with HPV58 and HPV45 (Fig. 3).

Figure 2.

Optimization for (A) E2 and (B) E6 genes of HPV16, and (C) E2 and (D) E6 genes of HPV18. The optimal Cq for HPV16 and HPV18 was selected at 63°C. HPV, human papillomavirus; Cq, quantification cycle.

Table II.

Analytical imprecision of the HPV16 (E2 and E6 genes) and HPV18 (E2 and E6 genes) measurement using multiplex quantitative polymerase chain reaction.

	HPV16 E2 gene		HPV16 E6 gene
Precision	Mean ± SD (CV%) Cq	Mean ± SD (CV%) Copy number	Mean ± SD (CV%) Cq	Mean ± SD (CV%) Copy number
Within-run (n=12)[a]
Low level[c]	32.41±0.18 (0.54)	961±127.1 (13.23)	33.67±0.17 (0.52)	1,084±149.1 (13.74)
High level[d]	23.11±0.08 (0.35)	952,490±59,040.35 (6.19)	24.59±0.10 (0.39)	1,097,119±81,784.1 (7.45)
Between-run (n=15)[b]
Low level[c]	32.61±0.20 (0.60)	973±141.36 (14.53)	34.26±0.19 (0.54)	929±137.45 (14.79)
High level[d]	23.37±0.14 (0.59)	911,140±91,626.87 (10.06)	25.04±0.14 (0.55)	1,040,597±112,028.74 (10.77)
	HPV18 E2 gene		HPV18 E6 gene
Precision	Mean ± SD (CV) Cq	Mean ± SD (CV) Copy number	Mean ± SD (CV) Cq	Mean ± SD (CV) Copy number
Within-run (n=12)[a]
Low level[c]	33.47±0.17 (0.52)	1,159±160.1 (13.81)	34.83±0.18 (0.52)	1,067±144.35 (13.52)
High level[d]	24.61±0.07 (0.29)	1,123,790±60,126.32 (5.4)	25.73±0.11 (0.44)	1,023,242±86,472.54 (8.45)
Between-run (n=15)[b]
Low level[c]	34.03±0.19 (0.57)	996±147.1 (14.76)	34.77±0.19 (0.53)	958±134.18 (14.00)
High level[d]	25.00±0.11 (0.43)	992,011±81,890.14 (8.25)	25.42±0.13 (0.51)	1,054,498±101,053.01 (9.58)

12 replicates per experiment

Triplicate in 5 independent experiments

103 copies of HPV16 and HPV18 DNA

106 copies of HPV16 and HPV18 DNA. HPV, human papillomavirus; Cq, quantification cycle.

Figure 3.

Analytical specificity of multiplex quantitative polymerase chain reaction. No cross reactivity was observed with 104 copies of HPV45 and HPV58. Uninfected HPV DNA was used as a negative control. HPV, human papillomavirus.

Evaluation of competitive effect of unequal HPV16 and HPV18 template concentrations in multiplex qPCR

To mimic the presence of HPV16 and HPV18 co-infection up to 1,000-fold difference in the same sample, measurement of mixed HPV16 and HPV18 DNA was compared in parallel with that of single HPV16 and HPV18. Different concentrations of HPV16 and HPV18 exhibited no effect on the quantification of E2 (P=0.319 and P=0.526, respectively) and E6 genes (P=0.347 and P=0.146, respectively), as demonstrated in Tables III and IV. Therefore, our established multiplex qPCR platform provided an accurate measurement for the presence of HPV16 and HPV18 co-infection.

Table III.

Competitive effect of unequal DNA copies between HPV16 and HPV18 on the measurement of HPV16 E2 and E6 genes using multiplex quantitative polymerase chain reaction.

HPVE2 gene				HPVE6 gene
Unequal mixture HPV16 and HPV18		Single HPV16		Unequal mixture HPV16 and HPV18		Single HPV16
Copy Ratio	Measured	Copy of	Measured copies	Copy Ratio	Measured	Copy	Measured copies
(HPV16:HPV18)	copies (Test)[a]	HPV16	(Control)[a]	(HPV16:HPV18)	copies (Test)[a]	of HPV16	(Control)[a]
106:103	945,367	106	1,026,287	106:103	979,497	106	1,039,326
106:104	901,450	106	915,772	106:104	900,328	106	913,776
106:105	921,335	106	897,221	106:105	914,673	106	911,439
106:106	979,241	106	980,160	106:106	1,039,326	106	964,189
105:103	89,554	105	90,112	105:103	87,556	105	89,341
105:104	95,749	105	103,356	105:104	91,586	105	98,371
105:105	96,968	105	105,065	105:105	97,092	105	103,174
105:106	94,159	105	92,332	105:106	112,708	105	100,329
104:103	8,832	104	9,149	104:103	11,219	104	10,439
104:104	8,836	104	10,626	104:104	9,175	104	9,303
104:105	8,913	104	9,319	104:105	8,641	104	10,322
104:106	11,347	104	10,861	104:106	10,989	104	9,018
103:103	1,143	103	1,200	103:103	1,039	103	1,153
103:104	867	103	912	103:104	907	103	865
103:105	881	103	1,122	103:105	848	103	973
103:106	848	103	901	103:106	846	103	908

Non-significant differences between test and control for HPV16 E2 (P=0.319; paired Student's t-test) and HPV16 E6 (P=0.347; paired Student's t-test). HPV, human papillomavirus.

Table IV.

Competitive effect of unequal copies between HPV16 and HPV18 on the measurement of HPV18 E2 and E6 genes using multiplex quantitative polymerase chain reaction.

HPVE2 gene				HPVE6 gene
Unequal mixture HPV18 and HPV16		Single HPV18		Unequal mixture HPV18 and HPV16		Single HPV18
Copy ratio	Measured copies	Copy of	Measured copies of	Copy ratio	Measured copies	Copy of	Measured copies of
(HPV18:HPV16)	of HPV18E2 (Test)[a]	HPV18	HPV18E2 (Control)[a]	(HPV18:HPV16)	of HPV18E6 (Test)[a]	HPV18	HPV18E6 (Control)[a]
106:103	1,135,661	106	1,075,620	106:103	1,015,464	106	1,107,241
106:104	1,097,976	106	1,181,450	106:104	1,107,170	106	1,125,579
106:105	877,385	106	1,135,265	106:105	911,742	106	893,768
106:106	1,092,086	106	993,922	106:106	965,678	106	1,040,114
105:103	104,611	105	96,998	105:103	90,546	105	94,778
105:104	101,164	105	113,785	105:104	97,952	105	113,933
105:105	105,464	105	101,469	105:105	110,175	105	87,996
105:106	102,760	105	113,715	105:106	87,492	105	107,587
104:103	9,129	104	8,789	104:103	10,201	104	9,746
104:104	9,550	104	9,443	104:104	9,732	104	10,395
104:105	8,639	104	9,988	104:105	10,403	104	8,665
104:106	10,700	104	9,812	104:106	8,910	104	10,550
103:103	912	103	1,120	103:103	937	103	1,056
103:104	936	103	1,078	103:104	910	103	938
103:105	849	103	831	103:105	901	103	833
103:106	896	103	1,168	103:106	869	103	1,063

Non-significant difference between test and control for HPV18E2 (P=0.526; paired t-test) and HPV18E6 (P=0.146; paired t-test). HPV, human papillomavirus.

Evaluation of physical status in clinical samples

Cut-off values for viral status were calculated as previously described by Wanram et al (18) and are shown in Table V. The E2/E6 ratio was identified as 0.78–1.10 and 0.85–1.18 for HPV16 and HPV18, respectively. An E2/E6 ratio of 0 was defined as the absolute integrated form, whereas E2/E6 >0 and less than the cut-off value was interpreted as the mixed form of episomal and integrated HPV (18). Comparisons of HPV16 and HPV18 copy numbers between the simplex and multiplex qPCR in CXCA samples are summarized in Table VI. No significant difference between simplex and multiplex qPCR was observed for HPV16 E2 and E6 (P=0.307 and P=0.288; paired t-test) and HPV18 E2 and E6 genes (P=0.396 and P=0252; paired t-test). The physical status obtained from multiplex qPCR was also compared with that from the simplex qPCR. The cut-off value for the episomal form of HPV16 (0.79–1.10) and HPV18 (0.85–1.18) was calculated as previously described for multiplex qPCR. Interpretation of physical status was similar in 95% (19/20) of cases between multiplex and simplex qPCR assays, and differed in one case, CX-1 (P=0.372; χ2 test; Table VI).

Table V.

Estimation of cut-off values (E2 and E6 ratios) for the interpretation of physical status using various concentrations of whole genome plasmid DNA from 104 to 106 copies.

		HPV16			HPV18
No.	Mixture of equal concentration of HPV16 and HPV18	E2 gene (copy)	E6 gene (copy)	E2/E6	E2 gene (copy)	E6 gene (copy)	E2/E6
1	103	1,115	1,103	1.01	1,153	1,233	0.94
2	103	1,115	1,185	0.94	1,019	1,210	0.84
3	103	908	1,044	0.87	1,193	1,211	0.99
4	103	1,168	1,169	1	1,155	1,134	1.02
5	103	1,038	1,047	0.99	1,146	1,179	0.97
6	104	8,882	10,814	0.82	10,760	9,858	1.09
7	104	8,789	10,438	0.84	9,187	9,408	0.98
8	104	9,019	10,379	0.87	8,703	9,062	0.96
9	104	9,233	9,268	1	8,513	8,220	1.04
10	104	9,117	9,337	0.98	9,687	11,622	0.83
11	105	94,603	105,819	0.89	110,462	107,970	1.02
12	105	108,055	108,928	0.99	107,699	102,882	1.05
13	105	88,246	100,449	0.88	106,389	98,269	1.08
14	105	87,938	105,370	0.83	114,166	112,276	1.02
15	105	106,246	100,978	1.05	109,971	113,442	0.97
16	106	1,098,531	1,059,641	1.04	1,086,195	991,320	1.1
17	106	926,410	982,153	0.94	1,097,976	996,524	1.1
18	106	912,783	1,076,185	0.85	1,085,161	927,645	1.17
19	106	1,020,080	931,143	1.1	1,086,195	1,005,826	1.08
20	106	940,240	997,234	0.94	1,146,435	1,142,588	1
			Mean	0.94		Mean	1.01
			SD	0.08		SD	0.08
			Mean ± 2SD	0.78–1.10		Mean ± 2SD	0.85–1.18

HPV, human papillomavirus; SD, standard deviation.

Table VI.

Comparison of viral copy number and physical status determination between multiplex qPCR (test) and simplex qPCR (control) in 20 cervical tissues samples harboring single and co-infection of HPV16 and HPV18.

	Simplex qPCR (Control)								Multiplex qPCR (Test)
	Copies/ng DNA		Physical status		Copies/ng DNA		Physical status		Copies/ng DNA		Physical status		Copies/ng DNA		Physical status
CXCA samples (HPV)	HPV16 E2	HPV16 E6	E2/E6 ratio (0.79)[a]	Interpretation	HPV18 E2	HPV18 E6	E2/E6 ratio (0.81)[b]	Interpretation	HPV16 E2	HPV16 E6	E2/E6 ratio (0.78)[c]	Interpretation	HPV18 E2	HPV18 E6	E2/E6 ratio (0.85)[d]	Interpretation
PCX 1 (16)	5,139	498,762	0.01	Mixed	–	–	–	–	4,196	470,492	0.01	Mixed	–	–	–	–
PCX 2 (16)	312	32,190	0.01	Mixed	–	–	–	–	203	30,613	0.01	Mixed	–	–	–	–
PCX 3 (16)	5,815	6,087	0.96	Episomal	–	–	–	–	5644.39	6196.23	0.91	Episomal	–	–	–	–
PCX 4 (18)	–	–	–	–	–	191,065	0	Integrated	–	–	–	–	–	153,540	0	Integrated
PCX 5 (18)	–	–	–	–	–	9,380	0	Integrated	–	–	–	–	–	11,897	0	Integrated
CX 1 (16)	118,584	157,522	0.75	Mixed	–	–	–	–	108,968	133,776	0.81	Episomal	–	–	–
CX2 (16)	30,353	26,353	1.15	Episomal	–	–	–	–	32,588	24,118	1.35	Episomal	–	–	–
CX 3 (16)	–	24,237	0	Integrated	–	–	–	–	–	26,814	0	Integrated	–	–	–
CX 4 (16)	38,041	44,122	0.86	Episomal	–	–	–	–	38,788	48,108	0.81	Episomal	–	–	–
CX 5 (16)	6,209	6,721	0.92	Episomal	–	–	–	–	6,879	7,395	0.93	Episomal	–	–	–
CX 6 (18)	–	–	–	–	15,649	45,455	0.34	Mixed	–	–	–	–	16,857	42,240	0.4	Mixed
CX 7 (18)	–	–	–	–	–	73,929	0	Integrated	–	–	–	–	–	70,912	0	Integrated
CX 8 (18)	–	–	–	–	74,237	56,271	1.32	Episomal	–	–	–	–	71,243	54,039	1.32	Episomal
CX 9 (16+18)	855	3,468	0.25	Mixed	326	521	0.63	Mixed	1,014	3,919	0.26	Mixed	205	420	0.49	Mixed
CX 10 (16+18)	39,350	123,420	0.32	Mixed	266	410	0.65	Mixed	36,000	155,429	0.23	Mixed	113	237	0.48	Mixed
CX 11 (16+18)	2,834,887	6,008,765	0.47	Mixed	4,090	3,087	1.32	Episomal	2,592,857	5,571,429	0.47	Mixed	3,064	2,414	1.27	Episomal
CX 12 (16+18)	4,122	14,979	0.28	Mixed	403	478	0.84	Episomal	3,305	17,360	0.19	Mixed	371	356	1.04	Episomal
CX 13 (16+18)	120	277	0.43	Mixed	219	415	0.53	Mixed	92	227	0.41	Mixed	197	339	0.58	Mixed
CX 14 (16+18)	9,124	24,510	0.37	Mixed	–	1,034	0	Integrated	8,733	20,600	0.42	Mixed	–	853	0	Integrated
CX 15 (16+18)	–	512,980	0	Integrated	–	981	0	Integrated	–	488,500	0	Integrated	–	741	0	Integrated
Internal control
HPV16; Caski	0	95,859	0	Integrated	0	0	0	–	0	91,791	0	Integrated	0	0	0	–
HPV18; HeLa	0	0	0	–	0	102,825	0	Integrated	0	0	0	–	0	110,275	0	Integrated
Whole plasmid HPV16	109,334	102,675	1.06	Episomal	–	–	–	–	106,445	105,988	1.00	Episomal	–	–	–	–
Whole plasmid HPV18	–	–	–	–	99,545	98,9655	1.01	Episomal	–	–	–	–	100,599	99,453	1.01	Episomal

The cut-off value of physical status interpretation for

HPV16

HPV18 using simplex qPCR; for

HPV16

HPV18 using multiplex qPCR. There was no significant difference in copy numbers between simplex and multiplex qPCR for HPV16 E2 gene (P=0.307; paired t-test); HPV16 E6 gene (P=0.288; paired t-test). There was no significant difference in copy numbers between simplex and multiplex qPCR for HPV18 E2 gene (P=0.396; paired t-test); HPV18 E6 gene (P=0.252; paired t-test). There was no significant difference in physical status from multiplex qPCR compared with simplex qPCR (P=0.372; χ2 test). Caski and HeLa cell lines containing 100,000 copies of E6 gene were used as internal standard for integrated HPV16 and HPV18/Whole plasmid HPV16 and HPV18 100,000 copies of E2 and E6 genes were used as internal standard for episomal HPV16 and HPV18. qPCR, quantitative polymerase chain reaction; HPV, human papillomavirus; PCX, pre-cancerous cervical tissues; CX, cervical cancer tissues.

Discussion

Multiple HR-HPV infection, particularly HPV16 and HPV18 co-infection, is now a concern due to its effects on cervical neoplasia development. The failure rate of treatment was previously reported to be increased by 5-fold in multiple infection (57%), compared with single infection (12%) (24). Therefore, a suitable risk assessment among patients with persistent multiple HR-HPV infection is required. To assess the risk, viral load and viral physical status may be used for cancer prognosis. In the present study, a one tube qPCR assay for HPV16 and HPV18 co-infection was successfully established with acceptable performance in terms of specificity, accuracy and precision.

Upon performing a literature search for multiplex qPCR of 4 genes in one tube, one study by Zhao et al (25) was identified. The authors reported detection of 4 viral DNAs: HPV16 (E6 gene), HPV18 (E6 gene), HSV1 and HSV2, with an improved detection limit at 10 copies compared with the present study. The difference in the detection limit may result from the different size of viral DNA standard. Small fragments of viral DNA (66–139 bp) were used as standard in the study undertaken by Zhao et al, whereas the whole HPV genome (10728–12267 bp) was used in the present study. Accordingly, the small size template has an advantage for amplification when compared with the whole genome. Therefore, our established technique better represents the real viral infection in a clinical setting. To resolve this limitation, more DNA template may be adjusted. The accuracy of interpretation, including possible cross reactivity, was verified using DNA of known viral status from cervical Caski and HeLa cell lines. HPV58 and HPV45 were selected for cross reactivity according to the top 4 common HR-HPV types (HPV 16, 18, 45 and 58), covering 90% of cases among Thai women (26). Accurate quantification was not only demonstrated via comparison with simplex qPCR, but it was also revealed that there was no competitive effect of an unequal mixed HPV DNA template from 10- to 1,000-fold (Tables III and IV).

Furthermore, the comparison of viral physical status between simplex and multiplex qPCR in clinical samples achieved 95% (19/20 samples) agreement in the results of viral physical status. The different physical status of CX1 was caused by the variation of E2/E6 ratio between simplex (0.75) and multiplex (0.81) which closed to the cut-off value for episomal form (0.79 and 0.78) resulting in discrepancy results as mixed form and episomal form, respectively. This limitation of accuracy occurred at values close to the cut-off values.

In conclusion, the increased incidence of HPV16 and HPV18 co-infection is a high-risk factor for CXCA progression in patients with persistent HR-HPV infection. Therefore, the successful development of multiplex qPCR for detecting HPV16 and HPV18 viral load and physical status in a single tube would provide a significant benefit in terms of cost effectiveness and shorter assay time in the clinic. To assess the potential of using this assay as a risk assessment for cancer progression in patients with single and co-infection with HPV16 and HPV18, a larger sample size with clinical outcome data should be included in future studies. In particular, pre-cancerous and early cancerous cases harboring high risk factors should be followed up frequently with monitoring of risk factors.