As soon as next year, the human papillomavirus (HPV) vaccine could transform clinical practice more than anything since the Pap smear was introduced 60 years ago.
A second large trial has shown extraordinary efficacy, and now 2 manufacturers, GlaxoSmithKline and Merck, are conducting late-phase clinical trials and working toward registering their vaccines for clinical use in 2006. Only last month, Merck and GSK signed an agreement that resolves their competing intellectual property claims—removing one more barrier to rapid commercialization.
But there are other important new developments that apply to practice now:
- Colposcopy, it appears, is far less reliable for identifying cervical intraepithelial neoplasia (CIN) 2,3 than we thought.
- The long-term risk of preterm delivery with loop electrosurgical excision procedures (LEEP) points to a need to counsel patients and consider all management options for women with CIN 1.
- The high rates of spontaneous regression of low-grade squamous intraepithelial lesion (LSIL) cytologic changes in young women are now better defined, and indicate colposcopy is not always needed.
Colposcopy not as sensitive as we thought
Pretorius R, Zhang W, Bellinson J, et al. Colposcopically directed biopsy, random cervical biopsy, and endocervical curettage in the diagnosis of cervical intraepithelial neoplasia II or worse. Am J Obstet Gynecol. 2004; 191:430–434.
We need to carefully follow up whenever colposcopy does not identify a CIN 2,3 lesion. This study also reinforces the need for diagnostic excisional procedures in women with an HSIL Pap result, and who are found after colposcopy to have CIN 1 or less (FIGURE 1).
Unfortunately, colposcopy is highly subjective. Accuracy depends on training and experience. Nevertheless, it is the standard of care for identifying CIN 2,3 and invasive cervical cancer in women with abnormal Pap results. Colposcopy was thought to be a sensitive but rather nonspecific method for identifying high-grade neoplasia. A 1998 comprehensive meta-analysis estimated that colposcopy had a weighted mean sensitivity for distinguishing normal tissue from abnormal tissue of 0.96 (95% confidence interval [CI], 0.95-0.97) and a weighted mean specificity of 0.48 (95% CI, 0.47-0.49).1 This means that colposcopy would miss a biopsy-confirmed cervical abnormality in only about 4% of patients. However, more recent follow-up studies have reported much higher false negative rates for colposcopy.
Pretorius and colleagues studied women enrolled in a cervical cancer screening trial conducted in Shanxi, China. The colposcopy in this study was performed by attending gynecologic oncologists who worked closely with a team of US-based gynecologic oncologists. The women in the study had biopsies taken of all areas classified as abnormal by colposcopy. In addition, random 4-quadrant cervical biopsies were obtained from colposcopically normal regions of the cervix.
A total of 364 women with a satisfactory colposcopy and biopsy-confirmed CIN 2 or greater lesions were identified. Even though all 364 women had a satisfactory colposcopic examination, only 57.1% of the women with biopsy-confirmed CIN 2 or worse were detected by the colposcopically-directed biopsy; the remaining 42.9% were detected by the random biopsies of colposcopically normal-appearing tissue. The lesions that were missed by colposcopy tended to be smaller than those identified by colposcopy and were more frequently CIN 2 rather than CIN 3 lesions.
This study also evaluated the role of endocervical curettage, and found that even among women with a satisfactory colposcopic examination, a significant proportion (5.5%) of cases of CIN 2,3 or worse were detected only by using endocervical curettage.
FIGURE 1 Repeat colposcopy and biopsy may reveal high-grade lesion
Low-grade cervical intraepithelial neoplasia (CIN 1) of the cervix. This young woman has a well-defined acetowhite lesion of her cervix that was diagnosed as a CIN 1 on cervical biopsy. In many such cases, repeat colposcopy and biopsy identifies an area of high-grade lesion that was missed at the initial colposcopy.
1. Mitchell MF, Schottenfeld D, Tortolero-Luna G, Cantor SB, Richards-Kortum R. Colposcopy for the diagnosis of squamous intraepithelial lesions: a meta-analysis. Obstet Gynecol. 1998;91:626-631.
LEEP raises risk of preterm birth
Sadler L, Saftlas A, Wang W, et al. Treatment for cervical intraepithelial neoplasia and risk of preterm delivery. JAMA. 2004;291:2100-2106.
We need to counsel women that LEEP will increase their risk for preterm premature rupture of membranes (PPROM) and preterm delivery. We must recognize that it is desirable to follow, rather than treat, biopsy-confirmed CIN 1, and to limit the depth of excision to 1 cm or less whenever possible.