Tailoring surgical treatment of cervical precancerosis

Summary

Objective:
To summarize new data which can help in decision on tailoring treatment of cervical precancerosis.

Design:
Review article.

Setting:
Department of Gynaecology and Obstetrics, First Faculty of Medicine, Charles University and Hospital Na Bulovce in Prague; Oncogynaecological Center, First Faculty of Medicine, Charles University and General University Hospital in Prague.

Results:
Precancerous lesions of the cervix are represented by squamous cervical intraepithelial neoplasias (CIN) and glandular adenocarcinomas in situ (AIS). The usual treatment of cervical precancerosis is conisation. However, some complications, particularly subsequent adverse pregnancy outcomes, follow all surgical treat-ments of cervix. The treatment could be postponed in women who wish to conceive and who suffer from CIN with a low risk of transformation to invasive cancer. The presence of modifying factors can help to stratify CIN lesions according to their malignant potential. The determination of detected HPV genotypes may help in this decision, because the fate of CIN 2/3 depends on the genotype of associated HPV infection. Cervical lesions associated with HPV 16, 18 or 45 are at a much higher risk of rapid progression to invasive cancers than lesions associated with other HR HPV genotypes. Surgical treatment of CIN 2/3 in women with a desire for future child-bearing can be postponed in cases non-associated with HPV 16, 18 and 45, on a case by case basis. Attempts are made to improve evaluation of the lesions by using biological and molecular markers, especially p16INK4a staining. Younger age, ongoing pregnancy, favourable colposcopic findings, negative p16^INK4a staining and immunocompetency are independent factors supporting the choice of conservative management. Adenocarcinoma in situ management substantially differs from the management of CIN.

Conclusion:
It is important both to assess all modifying factors correctly and to minimize any harm from unnecessary surgical treatment or overtreatment of cervical precancer lesions.

Keywords:
cervical intraepithelial neoplasia, adenocarcinoma in situ, conisation, tailoring treatment, human papillomavirus genotype, p16

Introduction

Precancerous lesions of the cervix are represented by squamous cervical intraepithelial neoplasias (CIN) and glandular adenocarcinomas in situ (AIS). Severe CIN (CIN 2/3, high-grade CIN) is a frequent disease, adenocarcinoma in situ (AIS) is much less common. Both of them, if untreated, may progress to invasive cancer. Cervical cancer is the second most common cause of death in women worldwide [3, 14, 49, 52]. In the Czech Republic the incidence of cervical cancer reaches the thirteenth highest rate in Europe with approximately 20 new cases per 100 000 women every year [48]. Nevertheless, the invasive cancer can be prevented in a majority of cases by an early treatment of cervical precancerosis.

The treatment methods and principles

The principle of CIN treatment is to eradicate the transformation zone (TZ) [36]. However, for adequate treatment, the entire lesion and not generally the entire TZ must be destroyed or excised.

Destructive methods include cold coagulation, cryotherapy, laser vaporisation, radical diathermy, and diathermocoalugulation [36]. The general term for ablative methods is conisation which means any kind of excision of the TZ [16, 36]. The ablative methods comprise cold knife cone biopsy (CKC), laser excision, and electrosurgical procedures (large loop excision of TZ - LLETZ, loop electrosurgical excision procedure-LEEP, needle excision of TZ - NETZ, and straight wire excision of TZ - SWETZ) [16, 36]. Randomized trials comparing different modalities show similar efficacy [20, 24, 34].

Surgical treatment of cervical precancerous lesions is an easy, cheap and effective, but can have serious negative consequences, especially for a subsequent pregnancy [2, 19]. However, a surgical treatment for precancerous cervical lesions is significantly associated with a spontaneous preterm birth following conization with the odds ratio (OR) varying from 1.52 to 4.4 at 24–27 weeks [1, 2, 19, 54].

Factors influencing decision on tailoring surgical treatment

At first, the expert should consider the age of woman. Young women have high regression rates of cervical disease [31]. Cancer risk remains generally low until the age of 25 [17, 31]. Fuchs et al. followed 40 adolescents and young women with biopsy-proven CIN 2 up to the age of 21. Only 8% of women from the study cohort had persistent CIN 2 or progressed to CIN 3 [10]. Current guidelines of the American Society for Colposcopy and Cervical Pathology recommend that young women till age of 24 years with CIN 2/3 should receive either treatment or observation for up to 12 months with both colposcopy and cytology in 6-monthly intervals [25]. When a histological diagnosis of CIN 2 is made in a young woman, observation is preferred, however the treatment is acceptable, too [25]. Treatment is recommended if colposcopy is inadequate, if CIN 3 is found, or if CIN 2 persists for 24 months [25]. Nevertheless, immediate treatment (ie, see-and-treat) of women 21–24 years old with a Pap smear result of HSIL is unacceptable without a previous biopsy [25]. No standard recommendations for pregnancy management following cervical surgery exist [15], so the desire for future pregnancy in young women should be the essential factor influencing the decision about treatment of precancerous lesion.

Expert colposcopy helps to assess the severity of the lesion. Especially very specific new colposcopic markers – ridge sign and inner border sign are frequently associated with CIN 2/3 and HPV 16 positivity [44]. Also the presence of the vascular pattern called „umbilication“ has 100% specificity, and 100% positive predictive value for the detection of underlying CIN 2/3 [43]. If colposcopy is unsatisfactory and other risk factors exist, a diagnostic excision should be considered because the occult high-grade lesion might be present [46].

The risk of progression of CIN 2/3 to invasive cancer during pregnancy is low and there is a high rate of postpartum spontaneous regression. The major reason for conservative management of precancerous lesions is the significant risk of complications (bleeding, premature rupture of membranes, preterm labor) after conization during pregnancy [7]. Pregnancy itself does not have an adverse effect on the prognosis of cervical lesions [57]. Moreover, a significant number of lesions can regress or even disappear postpartum. Yost et al. observed spontaneous regression in 69% from 153 women with previously histologically proven CIN 2/3 after puerperium, irrespective of the route of delivery [57].

It is clear that all CIN 2/3 lesions do not progress to invasive cancer. While the consensus for treatment of CIN 3 is unambiguous, the histological diagnosis of CIN 2 is extremely heterogeneous and has high inter-observer variation. There is the question of the real necessity to treat every confirmed CIN 2 lesion. Epidemiologic and molecular data suggest that CIN 2 is a diverse group with some cases similar to CIN 3 and some similar to CIN 1 with productive HPV infection. Approximately 40–43% of untreated CIN 2 and 30–32% of CIN 3 will regress spontaneously, 35% of CIN 2 and 56% of CIN 3 will persist, and only 5% of CIN 2 and 14 – 22% of CIN 3 will progress to invasive cancer [29, 35, 45, 27]. Women with high-grade CIN who do not have immediate diagnostic excision require close follow-up. But reliability of the follow-up is in such cases is limited [17] and management relies only on expert opinion [25]. According to both American and European guidelines, CIN 2 remains the consensus threshold for treatment, except in special circumstances [25, 36].

The incidence of AIS is low but rising [39]. AIS can be multifocal and discontinuous, so negative margins on an excision specimen do not provide assurance that the disease has been completely eradicated. Invasive cancer cannot be excluded without a diagnostic excisional procedure. For these reasons, total hysterectomy is preferred for women with AIS who have completed childbearing [25]. To repeat cytology is insufficiently sensitive to detect residual AIS or invasive adenocarcinoma compared with colposcopy and endocervical and endometrial explorative methods [18]. For women who wish to maintain fertility, observation after excisional procedure is an option, although it carries approximately 10% risk of persistent AIS and a small risk of cancer even when excision margins are negative [5, 8, 51]. If conservative management is planned and the margins of the specimen are involved or endocervical sampling obtained at the time of excision contains CIN or AIS, re-excision to increase the likelihood of complete excision is preferred [25].

HPV status

Persistent infection with high-risk (HR) human papillomavirus (HPV) types is necessary for the transformation of normal cervical epithelium to CIN and invasive cancer [3, 40, 41, 52]. The speed of the progression from CIN to cancer definitely depends on the infected HPV genotype. Tjalma et al. observed statistically significant difference in the time span between CIN 3 and SCC (squamous cell carcinoma) for HPV 45 (1 year), HPV 18(9 years), HPV 16 (15 years), HPV 33 (20 years), HPV 31 (23 years) and others (17 years) [49]. This supports the hypothesis that HPV 18 and 45 have a greater potential for rapid neoplastic transformation than the other types [4, 9, 37, 38, 49]. This finding could be used in decision on surgical treatment among women who desire to conceive. The conisation should be performed in cases of squamous epithelial cervical lesions related to HPV 16, 18, and 45, especially in association with HPV 45. On the contrary, lesions associated with HPV 31, 33, 35, and 52 can be observed and the surgical treatment can be postponed, on a case by case basis. The special attention should be paid to the genotype HPV 16 that presents a high risk due to its massive distribution and high incidence of reinfection. The speed of cancer induction by HPV 16 is rather low, with the time span of progression from CIN 3 to SCC of about 15 years. The conisation should be performed in case of CIN 3 associated with HPV 16 in case of CIN 2 the surgical treatment could be delayed if close follow-up with both colposcopy and cytology is performed.

Identification of true high-grade lesions

To improve the evaluation of lesions, biological and molecular markers in combination with other available techniques may be applied. Biomarkers sampled from punch biopsies are promising for the diagnosis of CIN 2/3 [50]. This promising strategy of studies of the dynamic behavior of CINs will hopefully allow identification of the regressive and non-regressive CIN 2/3 lesions and improve application of the patient-tailored treatment. On the other hand, there are no plasma or serum biomarkers with high sensitivity and specifi-city for the diagnosis and monitoring of cervical cancer [23].

Identifying true CIN 2/3, especially in young women, is a challenge that may be resolved with the aid of the most clinically validated marker p16^INK4a – the tumor suppressor that regulates transition from the G1 to the S phase of the cell cycle [53]. The high expression of the p16^INK4a is characteristic for dysplastic and neoplastic epithelium of the cervix [11]. Many studies have documented significant increase of the p16^INK4a expression in HSIL and SCC when compared to LSIL [32]. The detection of p16^INK4a can be definitively used as an adjunct to diagnose the precancerous and invasive cervical lesions [47].

Some studies have proposed that p16^INK4a is also a useful marker for predicting progression of CIN 1 lesions [57]. This information has recently been applied in clinical practice. Biopsy specimens with CIN 1 morphology but strong p16^INK4a expression should be up-staged as high-grade lesions; similarly CIN 2 morphology with basal p16^INK4a staining should be considered low-grade lesions and followed [32]. Immunostaining was also used in cases reported as CIN 1–2 or CIN 2–3 [55]. Futhermore, a recently published proposal for a new classification of cervical histopathology with two-tiered terminology distinguishes high-grade and low-grade squamous intraepithelial lesions. The classification is similar to the Bethesda System for the cervical cytology, which is familiar to all clinicians. Classification recommends p16^INK4a immunohistochemical staining in cases of morphological diagnosis of CIN 2. If the CIN 2 specimen is p16^INK4a positive, it is classified as HSIL; if p16^INK4a is negative, it is classified as LSIL [32].

Another marker, Ki-67, has been shown to be sensitive and specific for transforming HPV infection in the mature squamous epithelium [13]. A co-detection of p16^INK4a and Ki-67 in the same cells serves as an indicator of cell-cycle deregulation. It correlates with the HPV-induced oncogenic transformation and can therefore enhance overall accuracy and inter-observer agreement in diagnosing cervical biopsies. Positive results reliably identify LSIL which are likely to progress to CIN 2+ and can accurately categorize poorly reproducible CIN 2 as high-grade lesions [12, 33].

The oncogenic process in cervical cancer is initiated and mediated by the up-regulation of HPV E6/E7 oncoproteins, and thus, over-expression of these oncoproteins is a marker for an increased risk of cervical cancer [26, 42]. Therefore, the detection of the E6/E7 oncogene expression could be used as a specific a biomarker of cervical precancerosis and cancer [6]. Testing for HPV DNA reveals only the presence of the virus, whereas E6/E7 mRNA expression shows increased transcriptional activity of the cells, which is an important prognostic value of this test [22]. Repeated type-specific testing for HPV mRNA may identify young women with a persistent transforming infection being at increased risk for severe dysplasia [30] and the combination of mRNA E6/E7 HR HPV with colposcopic evaluation increases the colposcopy concordance with final histologic findings.

Productive viral infection is characterized by genome amplification and expression of late viral genes responsible for virion assembly, notably the capsid proteins (L1 and L2). This late gene expression is restricted to terminally-differentiated, superficial squamous epithelial cells.

Aberrant squamous cell differentiation that is considered a feature of a high-grade lesion is associated with a failure to express the capsid proteins and thus to complete the life cycle. Absence of L1 expression was a feature of progressive lesions in several studies using cytological specimens [21, 28]. The exact mechanism responsible for the loss of L1 and L2 expression in these lesions is not known. The expression of capsid proteins decreased with the increase of lesion grade (CIN 1 vs. CIN 2 vs. CIN 3 – p_trend < 0.0001) [56].

Conclusion

There are no relevant data supporting the assumption that all CIN 2/3 lesions will progress to cervical cancer. The presence of modifying factors can help to stratify CIN lesions according to their malignant potential. Long term follow-up studies of women with CIN 2/3 are needed, including HPV genotyping, p16^INK4a and other immunostains, cytogenetics, and use of other available markers, to guide the management of their cervical abnormalities. Development of a risk scoring system that incorporates cytology, histology, age of the patient, sexual history, smoking, immunological status, immunostains, HPV status and specific colposcopic features should help to create targetted management plans in women desiring childbirth with individual risks and benefits rather than using consensus guidelines. The treatment of CIN should be individualized in these women. As the number and sophistication of tools applied to cervical cancer prevention continue to increase, the complexity of exact evaluation and management promises to grow.

Acknowledgments

This paper has been supported by the IGA grant no. NT14079-3/2013 and PRVOUK-P27/LF1/1.

MUDr. Borek Sehnal

Onkogynekologické centrumGynekologicko-porodnická klinika

Nemocnice Na Bulovce

Budínova 2

180 81 Praha 8

e-mail: boreksehnal@seznam.cz