Microinvasive carcinoma of the cervix

This issue contains a short communication on microinvasive carcinoma of the cervix (Rambocas and Olah 2010, this issue.) This is not a new subject but the timing is useful because in recent years, there has been growing awareness that pre-term delivery might be a consequence of excision of cervical intraepithelial neoplasia (CIN): thus more patients with CIN in the future may be treated by ablative techniques to avoid subsequent obstetric complications.

FIGO staging of early invasive cervical cancer has evolved over the years but the latest (1994) is:

• Stage I: the carcinoma is strictly confined to the cervix; extension to the uterus corpus should be disregarded.

• Stage Ia: Invasive cancer identified only microscopically. All gross lesions seen with superficial invasion are Stage Ib. Invasion is limited to measured stromal invasion with a maximum depth of 5 mm and no wider than 7 mm diameter.

• Stage Ia1: measured invasion of stroma no greater than 3 mm in depth and no wider than 7 mm in diameter.

• Stage Ia2: measured invasion of stroma greater than 3 mm and no greater than 5 mm in depth and no wider than 7 mm diameter.

• Stage Ib: Cervical lesion confined to the cervix or preclinical lesions greater than stage Ia2 (Evans 1998).

The term ‘microinvasive’ refers to Stages Ia1 and Ia2.

The importance of the entity ‘microinvasion’ is that the patient does not require the same radical treatment as if she had a more invasive cancer. The chance of lymph node involvement is minimal and the overall prognosis better. Van Nagell et al. (1983) reported that of 52 patients whose cervical tumour invaded <3 mm, none of the 984 lymph nodes removed and examined contained tumour. If invasion was between 3 and 5 mm, three of 32 patients had involved nodes. Among 145 patients who were treated and found to have invasion <3 mm, only two returned with recurrent intraepithelial neoplasia; however, when invasion was 3–5 mm, there were three who subsequently returned with recurrent cancer, and two of them died. In the RCOG microinvasion study, there were 152 patients who fulfilled the established criteria; 142 had invasion up to 3 mm, and 10 had 3.1–5 mm, with one death in the study group (Morgan et al. 1996).

One of the hallmarks of early invasive cancer is neoangiogenesis or the development of additional blood vessels to provide for the growing tumour. Oddmund Koller (1963) described the appearance of vessels associated with cervical cancer in his monograph:

‘The squamous cell carcinomas deviated from the benign and precancerous areas in vascular patterns and surface contours … The vascular patterns presented a wide variety of pictures with varying size, shape or course … 

The most significant trend in the vascular pattern of squamous carcinomas was the lack of the normal dense and uniform distribution of fine capillaries, demonstrated by relatively enlarged avascular areas which could not be attributed to retention cysts or ordinary mosaic fields'.

Although Koller described these ‘atypical vessels’, his article also notes the avascular areas – this is a characteristic point that is often overlooked.

There are many older publications that associated atypical vessels with invasion. Sillman et al. (1981) reported that 82% of patients with atypical vessels had invasion, although they make the point that when the whole of the transformation was visible, only half of the patients with microinvasion had atypical vessels seen. Noda (1981) described that 67% of patients with Stage Ia1 lesions had atypical vessels, but this appears to be the highest percentage described. Benedet et al. (1985) in their series of 180 Stage Ia patients made the point that a high percentage (42%) of patients with microinvasion had unsatisfactory colposcopic examinations, and that colposcopy was less sensitive in detecting microinvasive lesions than in detecting occult carcinoma. Kolstad (1989) reported that colposcopy predicted microinvasion in only 32.6% of his large series.

There was a series of reports (Wongtra-Ngan et al. 1988; Kolstad 1989; Paraskevaidis et al. 1992; Bowen et al. 2000; Cairns and Cruickshank 2007) that the majority of women with microinvasion were not suspected from referral cytology, colposcopy appearances and only recognised when the transformation zone was excised. Therefore, what will be the impact of using ablative techniques when avoiding cervical excision?

Earlier studies on the transformation zone (Pryzbora and Plutowa 1959; Burghardt and Ostor 1983) described the topography of the transformation zone and identified metaplastic followed by neoplastic involvement of the cervical glands or crypts. Anderson and Hartley (1980) described that the mean depth of crypt involved with cervical intraepithelial neoplasia was 1.24 mm, and that the mean plus 3 standard deviations would include a depth of 3.8 mm; thus, destruction to this depth would eradicate 99.7% of involved crypts. Similar findings were reported by Abdul-Karim et al. (1982) that the mean depth of involvement with CIN3 was 1.35 ± 1.15 mm, and to eradicate 99.7% of lesions, it would be necessary to destroy tissue up to 4.8 mm depth.

Therefore, if selective ablation of CIN destroys to a depth of 5 mm, the failure to recognise microinvasion in areas of CIN probably does not have clinical repercussions (Wongtra-Ngan et al. 1988). However, if the entire transformation zone is not seen, i.e. the colposcopic examination is unsatisfactory, or there are cytological or colposcopic (atypical vessels plus avascular areas) features to raise the possibility of invasion, excision must be performed. Equally, those clinicians performing ablative or destructive treatments of CIN must have the experience or techniques to judge depths of 5 mm; failure to destroy to adequate depths will run an increased risk of subsequent cancers (Sevin et al. 1979; Townsend et al. 1981).

Declaration of interest: The author reports no conflicts of interest. The author alone is responsible for the content and writing of the paper.