To study the correlation between HPV detection using self-collected urine samples, brush-based self-samplers and physician-taken smears and to compare HPV detection in morning first-void urine to HPV detection in first-void urine from any hour during the day.
Women referred to a colposcopy clinic after an abnormal PAP-smear result were sent a device (Colli-Pee™, Novosanis, Wijnegem, Belgium) to collect first void-urine on the first urine of the morning of their colposcopy. A second first-void urine sample was collected during the visit, together with a physician-taken cervix smear, a brush-based self-sample (Evalyn brush™, Rovers Medical Devices B.V., Oss, The Netherlands) and a colposcopy directed biopsy. Urine samples, smear and brushes were tested with the analytically sensitive SPF10-DEIA-LiPA25version1 assay and the clinically validated GP5+/6+-EIA, with LMNX genotyping. Histological material was assessed by a local pathologist.
Morning (U1) and afternoon (U2) urine samples, physician-taken smear (PTS) and brush based self-samples (SS) from 91 patients were analysed. All CIN3 lesions (N=6) were hrHPV positive in PTS, SS, U1 and U2 with both the SPF10-assay and the GP5+/6+-assay (Table 1).
The sensitivity for CIN2+ detection in PTS, SS, U1 and U2 with the SPF10 was 96.4%, 92.9%, 92.9% and 96.4% respectively (p>0.05). With the GP5+/6+ assay, the sensitivity was 78.6% in PTS, 82.1% in SS, 92.9% in U1 and 85.7% in U2 (p>0.05).
On genotype level, a substantial to almost excellent agreement was found between all samples (kappa 0.663-0.912), for both SPF10 and GP5+6+. When comparing the genotypes found in PTS to those found in urine samples, 12.6% of the samples were discordant when tested with SPF10 and 19.8% of the samples when tested with GP5+/6+. For the comparison between SS and urine, discordant results were found in 10.0% and 14.3% of the samples, respectively. When comparing PTS to SS, 11.0% of the samples were discordant with SPF10 and 16.5% with GP5+/6+. Between U1 and U2, 3.2% were discordant with SPF10 and 6.6% with GP5+/6+.
| Histological diagnosis | PTS | PTS | SS | SS | U1 | U1 | U2 | U2 |
| SPF10 | GP5+/6+ |
SPF10 |
GP5+/6+ |
SPF10 |
GP5+/6+ |
SPF10 |
GP5+/6+ | |
| Negative (46) | 56.6% | 47.8% | 56.5% | 50.0% | 58.7% | 45.7% | 63.0% | 52.2% |
| CIN1 (17) | 82.4% | 82.4% | 82.4% | 76.5% | 82.4% | 76.5% | 82.4% | 70.6% |
| CIN2 (22) | 95.5% | 72.7% | 90.9% | 77.3% | 90.9% | 90.9% | 95.5% | 81.8% |
| CIN3 (6) | 100% | 100% | 100% | 100% | 100% | 100% | 100% | 100% |
| Total | 67 | 58 | 66 | 59 | 60 | 60 | 70 | 60 |
CIN2+ detection using HPV testing in first-void urine seems feasible, with sensitivity similar to the sensitivity of the physician-taken smears and brush-based self-samples. No advantage in testing morning first-void urine over first-void urine from later during the day was found.