Review

Med Princ Pract 2008;17:351–364 Received: November 7, 2007

Revised: February 4, 2008
DOI: 10.1159/000141498

Cervical Cancer: Screening and

Therapeutic Perspectives
Rengaswamy Sankaranarayanan a Somanathan Thara b Pulikottil Okkuru Esmy c
Partha Basu d
a
Screening Group, World Health Organization-International Agency for Research on Cancer, Lyon, France;
b
Department of Pathology, Regional Cancer Centre, Trivandrum, c Department of Radiotherapy and
Clinical Oncology, Christian Fellowship Community Health Centre, Ambillikai, and d Department of
Gynaecological Oncology, Chittaranjan National Cancer Institute, Kolkata, India

Key Words affordability and capacity for initiating and sustaining vac-
Cervical cancer, screening and therapy cination and screening programmes are critical factors in
cervical cancer control. On the other hand, an informed
practitioner can utilize the multiple opportunities in routine
Abstract primary care interactions for prevention, screening, early de-
Cervical cancer is a major cause of mortality and premature tection and prompt referral for treatment.
death among women in their most productive years in low- Copyright © 2008 S. Karger AG, Basel
and medium-resourced countries in Asia, Africa and Latin
America, despite the fact that it is an eminently preventable
cancer. While cytology screening programmes have resulted Introduction
in a substantial reduction of cervical cancer mortality in de-
veloped countries, they have been shown to have a wide Cervical cancer is an important global public health
range of sensitivity in most routine settings including in de- problem. It accounted for an estimated 493,000 incident
veloping countries. Although liquid-based cytology im- cases, 1.4 million prevalent cases and 273,000 deaths in
proves sample adequacy, claims on improved sensitivity re- the world around the year 2002, constituting approxi-
main controversial. Human papillomavirus testing is more mately 8% of the global burden of cancer among women
sensitive than cytology, but whether this gain represents (table 1) [1]. Developing countries accounted for four
protection against future cervical cancer is not clear. Recent- fifths of this global burden, reflecting the grim reality of
ly, in a randomized trial, the use of visual inspection with 4% the lack of effective control measures in many high-risk
acetic acid was shown to reduce cervical cancer incidence countries. It is a major cause of mortality and premature
and mortality. Cryotherapy and large loop excision of the death among women in their most productive years in
transformation zone are effective and safe treatment meth- low- and medium-resourced countries in Asia, Africa
ods for cervical intraepithelial neoplasia. The clinical stage of and Latin America, despite the fact that it is an eminent-
cancer is the single most important prognostic factor and ly preventable cancer. If effective prevention interven-
should be carefully evaluated in choosing optimal treatment tions are not implemented, over 1 million new cervical
between surgery and radiotherapy, with or without chemo- cancer cases will be diagnosed annually by the year 2030.
therapy. At the public health level, health care infrastructure, While early detection of asymptomatic precancerous le-

© 2008 S. Karger AG, Basel Dr. R. Sankaranarayanan

1011–7571/08/0175–0351$24.50/0 Screening Group, World Health Organization-International Agency for
Fax +41 61 306 12 34 Research on Cancer, 150 cours Albert Thomas
E-Mail karger@karger.ch Accessible online at: FR–69372 Lyon Cedex 08 (France)
www.karger.com www.karger.com/mpp Tel. +33 472 73 85 99, Fax +33 472 73 85 18, E-Mail sankar@iarc.fr
Table 1. Cancer of the uterine cervix: incident cases, deaths and per 100,000 in many countries. Rates lower than 7 per
5-year prevalence in 18 world regions in 2002 100,000 women are observed in middle-eastern coun-
tries, while these are lower than 10 per 100,000 women in
Region Cervix cancer burden
most developed countries [1, 3]. A large variation in sur-
cases deaths 5-year vival from cervical cancer is observed among countries
prevalence due to the differences in clinical stages at presentation
World 492,800 273,200 1,409,200 and the level of development of cancer-related health ser-
More developed countries 83,400 39,500 309,900 vices in different countries. Five-year survival rates of less
Less developed countries 409,400 233,700 1,099,300 than 25% are reported for black patients in Uganda [4]
Eastern Africa 33,900 27,100 57,200 and Zimbabwe [5]; survival ranged between 30 and 50%
Middle Africa 8,200 6,600 13,900
in Cuba, India, and Philippines, 50 and 60% in Thailand
Northern Africa 8,100 6,500 14,000
Southern Africa 7,600 4,400 13,100 and mainland China [6], and 65% in Singapore [7]. Rates
Western Africa 20,900 16,700 35,700 ranged between 60 and 75% in developed countries
Caribbean 6,300 3,100 18,400 [8, 9].
Central America 17,100 8,100 49,300 Estimated age-adjusted cervical cancer mortality rates
South America 48,300 21,400 139,200 ranged between 3 and 8 per 100,000 women in most de-
Northern America 14,600 5,700 58,200
Eastern Asia 61,100 31,300 191,900 veloped countries and 10–25 per 100,000 women in most
South-Eastern Asia 42,500 22,500 132,500 developing countries [1]. The high mortality in develop-
South Central Asia 157,700 86,700 446,100 ing countries is due to advanced clinical stage at presen-
Western Asia 4,400 2,100 13,700 tation and to the fact that a significant proportion of pa-
Eastern Europe 30,800 17,100 107,700 tients does not avail of or complete prescribed courses of
Northern Europe 5,600 2,800 21,100
Southern Europe 10,600 4,100 40,900 treatment due to deficiencies in treatment availability, ac-
Western Europe 12,700 5,600 49,200 cessibility and affordability. There is reason to suspect
Oceania 2,000 800 6,500 that the burden of disease in some of the high-risk coun-
tries, particularly in sub-Saharan Africa, is underesti-
mated given the inadequacy of diagnostic and treatment
services and lack of reliable cancer information systems.

sions by screening and their effective treatment lead to

the prevention of invasive cervical cancer and premature Natural History of Cervical Cancer
death from it, the fact that cervical cancer is caused by
persistent infection by one or more of the 15 oncogenic Compared to many other human cancers, the natural
human papilloma virus (HPV) types [2] provides the ex- history of cervical neoplasia is much better understood.
citing opportunity for prevention through vaccination. The direct precursors to invasive squamous carcinoma
We will briefly discuss the role of screening for preneo- are the high-grade cervical intraepithelial neoplasia (CIN
plastic disease, early clinical diagnosis and treatment of grades 2 and 3). One third to a half of these may progress
invasive cancer in the control of cervical cancer, with to invasive cervical cancer over a period of 5–15 years,
particular reference to low-resource countries where lack while most low-grade CIN (CIN 1) regress spontaneous-
of effective screening programmes, early diagnosis and ly and only 10–15% persist or progress to high-grade le-
diagnostic/treatment facilities have resulted in high cer- sions. Adenocarcinoma in situ is the precursor lesion for
vical cancer death rates. invasive adenocarcinoma. Persistent infection with one
or more of the oncogenic types of HPV causes cervical
neoplasia [2]. Over 99% of cervical cancer cases and their
Cervical Cancer Incidence and Mortality precursors are caused by persistent infection with HPV,
an infection of the surface epithelium that is mostly
There is a more than eightfold difference between the asymptomatic. Women are mostly infected with HPV in
highest and lowest incidence rates of cervical cancer their teens, twenties, or early thirties and it is estimated
worldwide (fig. 1) [1, 3]. In sub-Saharan Africa, Central that 50–80% of women may be infected after sexual debut
and South America, South Asia and South-East Asia, age- in their lifetime. Most HPV infections are transient and
standardized incidence rates of cervix cancer exceed 25 80% of them resolve within 2 years, but some women in-

352 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu

 ASR (World) per 100,000 women
0 5 10 15 20 25 30 35 40 45

Eastern Africa
Southern Africa
Melanesia
Caribbean
Central America
Western Africa
South America
Middle Africa
Polynesia
South Central Asia
Less developed countries
South-Eastern Asia
World
Eastern Europe
Northern Africa
Southern Europe
More developed countries
Western Europe
Micronesia
Northern Europe
Northern America
Eastern Asia Incidence
Mortality
Australia/New Zealand
Fig. 1. Incidence and mortality rate of cer- Western Asia
vical cancer in selected regions. ASR =
Age-standardized incidence rates.

fected with one or more of the high-risk types develop genotypes that cause cervical cancer; the frequency of
persistent infections. Persistent genital HPV infection HPV in cervical cancer specimens ranged from 86 to 94%
can lead to high-grade CIN (CIN 2–3) and adenocarci- in different regions of the world [11]. Almost the same
noma in situ, which, left undetected and untreated, can estimates were found in a meta-analysis of 85 published
lead to both invasive squamous cell carcinomas and ad- studies [12].
enocarcinoma. Of the more than 100 HPV types identi-
fied, one or more of the 15 so-called high-risk types (HPV
16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, 73 and 82) Prevention by Vaccination
are associated with cervical neoplasia [2]. In a pooled
analysis of 3,085 cases of cervical cancer in 11 studies The estimated HPV 16/18 positive fraction in cervical
from 25 countries, the overall HPV prevalence was 96% cancer and high-grade CIN is approximately 70 and 50%,
where HPV 16 and 18 accounted for 70% and the next six respectively, which equates to 350,000 potentially pre-
most common types accounted for 19% of cases [10]. In a ventable cervical cancers annually by vaccination against
meta-analysis involving 14,595 cases of invasive cervical HPV 16 and 18 [1, 12]. Thus, there is a vast potential to
cancer, 87% contained one or more of the oncogenic HPV reduce HPV-related morbidity and mortality by HPV

Screening and Therapy of Cervical Med Princ Pract 2008;17:351–364 353

Cancer
vaccination. Currently, monovalent (HPV 16), bivalent nized and effective population-based cervical cancer
(HPV 16, 18) and quadrivalent (HPV 6, 11, 16, 18) HPV screening programmes have not yet been implemented in
L1 virus-like particle vaccines have been developed and most developing countries due to several barriers, such as
evaluated [13–21]. The results from these studies indicate, competing health care priorities, and limited, under-
with remarkable consistency, that a regimen of three in- staffed, under-resourced and overstretched primary
tramuscular injections of HPV vaccine offers HPV-naïve health care facilities [23]. Needless to say, cancer diagnos-
women a very high level of protection (⬃99%) from infec- tic treatment and palliative care services are even more
tions and CIN associated with the HPV types included in limited in many of these countries.
the vaccine. The vaccines were safe and well tolerated
with relatively few side effects [13–21]. While HPV vac-
cination holds great promise, there are still several chal- Screening Methods for Cervical Neoplasia
lenges that need to be resolved before it can be widely
implemented in high-risk developing countries [22]. Conventional cervical cytology, liquid-based cytology
These challenges include current high costs of the vac- (LBC), HPV testing and visual screening after applica-
cines, affordability, feasibility, acceptability, logistics of tion of acetic acid (VIA) or Lugol’s iodine (VILI) are the
vaccine delivery (in view of the need for three doses currently available tests for the early detection of CIN.
spread over 6 months, improved strategies and vaccine The most widely used and evaluated screening test is con-
platforms to reach out to pre- or early-adolescent girls), ventional cytology and, in recent years, the other tests
long-term immunogenicity and efficacy in preventing have been increasingly evaluated in different settings.
cervical neoplasia, cross-protection against HPV types The accuracy and efficacy of these tests in preventing
not targeted by the vaccine antigens and the efficacy of cervical cancer will be discussed briefly.
different, more logistically feasible dose regimes in in-
ducing and maintaining immunogenicity and long-term Conventional Cytology
protection against cervical neoplasia. Prophylactic vac- Cytology (Pap smear) screening involves collection of
cination is likely to provide important future health gains cervical cell samples from the cervical epithelium using
if vaccination is offered to girls before onset of sexual ac- a wooden spatula or a brush, preparation and fixation of
tivity. However, cervical screening should still be contin- the smear by a doctor or a nurse followed by staining,
ued because the risk of being infected with the oncogen- reading and reporting of the results by a cytotechnician
ic HPV types or of acquiring CIN due to an HPV type and a cytopathologist. Cytology requires a laboratory in-
other than 16 or 18 remains. frastructure, with internal and external quality control
measures to process slides and microscopy, and a system
to communicate the results to those concerned. High-
Screening for Cervical Cancer quality training, continuing education, and proficiency
testing of personnel are essential to ensure reliable and
Screening involves application of a relatively simple, efficient testing.
inexpensive test to a large number of asymptomatic wom- In most routine settings, cytology has been shown to
en in order to classify them as likely or unlikely of having have a wide range in sensitivity in detecting cervical neo-
the disease of interest. The screen-positive women are plasia. The sensitivity to detect CIN 2 and 3 lesions ranged
subsequently investigated to rule out cervical neoplasia from 47 to 62% and the specificity from 60 to 95% in re-
and those confirmed with disease are treated. The objec- views of several studies [24, 25]. There have been several
tive of cervical screening is to prevent invasive cervical cross-sectional studies in developing countries assessing
cancer by detecting and treating women with high-grade the accuracy of cytology, in which the sensitivity varied
CIN 2 and 3 lesions and the effectiveness of screening is from 31 to 78% and the specificity from 91 to 96% (ta-
evaluated by the extent of reduction in cervical cancer ble 2) [26]. Verification bias was minimized in some of
incidence and mortality following screening. The critical these studies by providing reference standard investiga-
components of successful cervical screening are high tions to all participants.
coverage of target women with accurate, quality-assured Large-scale population-based cytology screening pro-
screening tests, screen-positive women with diagnostic grammes have resulted in a dramatic reduction in the
investigations and women with confirmed cervical neo- burden of cervical cancer in the past five decades in the
plasia with treatment and follow-up care. However, orga- developed countries of Europe, North America, Japan,

354 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu

Australia and New Zealand by early detection and effec- Table 2. Characteristics of cervical screening tests for detecting
tive treatment of cervical precancerous lesions [23, 27]. CIN 2 and 3 lesions in cross-sectional studies
The marked reduction in the incidence of and mortality
Test Sensitivity, % Specificity, %
from cervical cancer before and after the introduction of
cytology screening in developed countries has been in- Cytology 31–78 91–96
terpreted as strong non-experimental, observational evi- HPV testing 61–100 62–96
dence for the effectiveness of cytology screening pro- VIA 37–95 49–97
grammes [27]. Following the introduction of widespread VILI 44–92 75–85
cytology screening in developed countries since the
1960s, cervical cancer mortality rates have declined at a
remarkable rate. Cervical cancer incidence has been re-
duced by as much as 80% where the cytology screening one single or two visits to complete the screening and di-
quality, coverage and follow-up of women are high. The agnosis/treatment processes [23, 26, 27]. This has also led
highest reduction in cervical cancer incidence was in the to the reorganization of existing cytology programmes
30–49 age groups where the focus of screening was the and more effective utilization of resources in some coun-
most intense. Organized screening with systematic call, tries such as Brazil, Chile, Costa Rica, South Korea, Mex-
recall, follow-up and surveillance systems have shown ico, Singapore, Thailand and Uruguay. Reorganization of
the greatest effect (e.g. Finland, Iceland), while using few- the programme in Chile has been associated with decline
er resources than the less organized programmes (e.g. in cervical cancer mortality in recent years [28]. The in-
USA, France). cidence and/or mortality rates are currently declining in
Establishing quality-assured cytology screening pro- Costa Rica, Hong Kong, Singapore, Taiwan, Mexico,
grammes with national coverage is beyond the capacity Uruguay and South Korea due to widespread screening
and resources of many developing countries in view of in the health services.
the infrastructure for testing, trained personnel for read-
ing, quality assurance and the organization required. Cy- Liquid-Based Cytology
tology screening has failed to reduce cervical cancer bur- LBC relies on a uniform thin layer of cervical cells
den by any great extent in Latin American countries in- without debris prepared from processing a fluid medium
cluding Cuba, Brazil, Mexico, Peru, and Colombia [23]. containing the cervical cells. The advantages of LBC in-
The main reasons for the lack of success in these coun- clude an increased possibility of a more representative
tries were a combination of suboptimal cytology testing, and complete transfer of cervical cells from the sampling
lack of quality assurance, poor coverage of women at risk device to the slide and improved microscopic readability
and inadequate follow-up of screen-positive women with due to the elimination of problems such as poor fixation,
diagnosis and treatment. While poor-quality cytology is air-drying artifact, uneven thickness of the cellular
a reflection of several challenges in providing quality- spread, debris due to blood and inflammatory cells, and
assured testing, the lack of coverage for diagnosis and overlapping of cells. Cell suspension remaining after the
treatment is related to the inadequate health care infra- preparation of the smear may be used for additional test-
structure, human resources and programme logistics. ing procedures such as HPV testing. It is a more expen-
Cytology is a resource-intensive technology and, unfor- sive test than conventional cytology and requires addi-
tunately, it is not possible to commit sufficient financial tional instrumentation to prepare the smears. LBC is re-
and human resources to fulfil the optimal requirements. ported to improve sample adequacy and increase the
These would be for collection of cervical cells, slide prep- sensitivity of cervical cytology in comparison with con-
aration, staining, reading and reporting, as well as qual- ventional cytology [25, 29].
ity control measures, to ensure good-quality cytology In a recent review of 56 studies, LBC and convention-
with optimal accuracy in low-resource countries. al cytology were compared in terms of the percentage of
A critical appraisal of reasons for the failure or subop- slides classified as unsatisfactory in each cytology catego-
timal performance of cytology screening and the diffi- ry and the accuracy of detection of high-grade disease
culties in organizing cytology screening in low- and me- [30]. Data were examined for studies overall and in strata
dium-resourced countries have prompted the search for, to examine the effect of study quality on results. The me-
and evaluation of, alternative screening tests such as VIA, dian difference in the percentage of unsatisfactory slides
VILI and HPV DNA testing and paradigms that require between LBC and conventional cytology was 0.17%. The

Screening and Therapy of Cervical Med Princ Pract 2008;17:351–364 355

Cancer
classification of high-grade squamous epithelial lesions specificity 0.95 (95% CI = 0.94–0.96). For women over 30
varied according to study quality (p = 0.04), with conven- years, the sensitivity of HC2 was 94.8% and the specific-
tional cytology classifying more slides in this category ity 86.0%. The combination of HC2 and cytology has the
than did LBC in high-quality studies (n = 3) only. In me- highest sensitivity and lowest specificity. The sensitivity
dium-quality (n = 30) and high-quality studies, LBC clas- of HPV testing reported from developing country set-
sified more slides as atypical squamous cells of unknown tings has been somewhat lower than that reported from
significance (ASCUS) than did conventional cytology studies in developed countries [26].
when compared with low-quality studies (n = 17; p = In randomized trials, HPV testing has greater sensi-
0.05). This review does not lend support to claims of bet- tivity for the detection of CIN as compared with Pap test-
ter performance by LBC. ing [34–36]. In a randomized trial in Sweden, 12,527
In a randomized trial in Italy, LBC showed no signifi- women aged 32–38 years were randomly assigned to have
cant increase in sensitivity for CIN 2 or worse lesions (rel- HPV test plus cytology (intervention group) or a Pap test
ative sensitivity 1.17, 95% confidence interval 0.87–1.56), alone (control group) [34]. The relative frequency of CIN
whereas the positive predictive value was reduced (rela- 2 or 3 or cancer detected at enrollment in the intervention
tive positive predictive value vs. conventional cytology group was 51% greater than the proportion of women in
0.58, 0.44–0.77) [31]. LBC detected more lesions of grade the control group. At subsequent screening examinations
1 or more (relative sensitivity 1.68, 1.40–2.02), with a larg- over 4 years, the proportion of women in the intervention
er increase among women aged 25–34, but did not detect group who were found to have CIN 2 or 3 lesions or can-
more lesions of grade 3 or more (relative sensitivity 0.84, cer was 42% less and the proportion with CIN 3 lesions
0.56–1.25). The relative frequency of women with at least or cancer was 47% less than the proportions of control
one unsatisfactory result was lower with LBC (0.62, 0.56– women. In a randomized trial in Canada involving 10,154
0.69). women aged 30–69 years, the sensitivity of HPV testing
It is not feasible to implement LBC in many low-re- for CIN 2 or 3 was 94.6%. These used the conservative
source settings. Although some countries have changed definition based on confirmation by histological and, in
to LBC for cervical screening, controversy remains. The addition, loop electrosurgical excision procedure (LEEP)
impact of LBC on cancer incidence and mortality re- investigations. On the other hand, the sensitivity of Pap
mains to be established, as does its cost-effectiveness. testing was 55.4% (95% CI, 33.6–77.2; p = 0.01) [35]. The
specificity was 94.1% for HPV testing and 96.8% for Pap
HPV Testing testing. The sensitivity of both tests used together was
The fact that cervical neoplasia is caused by persistent 100%, and the specificity was 92.5%. In a randomized
infection with oncogenic types of HPV has led to the trial in the Netherlands, 8,575 women aged 29–56 years
evaluation of HPV testing as a primary screening test for were assigned to receive combined cytological and HPV
cervical neoplasia. HPV testing is the most objective and DNA testing and 8,580 women (control group) conven-
reproducible of all currently available cervical screening tional cytological testing only. CIN 3 or worse lesions de-
tests. The sensitivity of HPV testing in detecting CIN 2 tected at baseline in the intervention group was 70% more
and 3 lesions varied from 66 to 100% and the specificity than in the control group (68 vs. 40, p = 0.007) [36]. The
varied from 62 to 96% in several cross-sectional studies number of CIN 3 or worse lesions detected in the subse-
(table 2) [26, 32, 33]. In a recent meta-analysis of 25 stud- quent round was lower in the intervention group than in
ies comparing HPV testing to cytology for the detection the control group (24/8,413 vs. 54/8,456, 55% decrease,
of high-grade CIN in primary cervical cancer screening, 95% CI 28–72; p = 0.001). The number of CIN 3 or worse
the pooled sensitivity of HPV testing by hybrid capture 2 lesions over the two rounds did not differ between groups,
(HC2) test, polymerase chain reaction, cytology (ASCUS implying that HPV DNA testing leads to earlier detection
or worse) and cytology (LSIL or worse) was 90, 80.9, 72.7 of CIN 3 or worse lesions.
and 61.6%, respectively, and the pooled specificity was Although self-sampling for HPV DNA testing seems
86.5, 94.7, 91.9 and 96.0%, respectively [32]. The ratio of to be a viable screening option and potentially promising
the sensitivity of HC2 to cytology (ASCUS) was 1.25 (95% for use in under-resourced areas or for women who are
CI = 1.20–1.29), and the corresponding specificity ratio reluctant to participate in screening programmes, evi-
was 0.97 (95% CI = 0.96–0.98). The ratio of the sensitiv- dence supporting it is limited [37]. Further definitive re-
ity of combination of HC2 and cytology (ASCUS) to HC2 search is needed to clarify the use of self-sampling for
alone was 1.05 (95% CI = 1.04–1.06) and the ratio of the HPV DNA testing for the purpose of increasing screen-

356 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu

ing rates, especially in women who are never or seldom lowing VIA allow diagnostic investigations and/or treat-
screened. ment in the same session as screening. A range of person-
In low-resource settings, where repeated testing of nel including doctors, nurses, midwives, and paramedi-
women at risk for cervical neoplasia may not be feasible, cal health workers can be rapidly trained in providing
HPV testing may provide an objective method of identify- VIA in short training courses of 4–10 days [41]. A wide
ing and investing the limited resources on women at risk range of teaching materials is now available for training
for disease [26]. However, it is currently more expensive personnel in carrying out VIA competently [40, 42, 43].
(20–30 USD) than other screening tests and requires so- However, it is a subjective test that suffers from high false-
phisticated laboratory infrastructure including testing positive rates and low to moderate specificity and repro-
equipment, storage facilities for samples and trained tech- ducibility. Quality assurance procedures for VIA are yet
nicians. Further developments in terms of less expensive to be standardized and assuring consistent high perfor-
testing and less sophistication in infrastructure and equip- mance can be challenging under field conditions and re-
ment requirements are essential to make HPV testing fea- quires constant monitoring and frequent retraining of
sible in low-resource settings. Efforts are now underway to test providers.
develop simple, affordable, rapid and accurate HPV testing The sensitivity of VIA to detect CIN 2 and 3 lesions
methods for use in low- and medium-resource settings. and invasive cervical cancer varied from 37 to 95% and
In summary, compared to cytology, HPV testing is the specificity varied from 49 to 97% in several cross-sec-
substantially more sensitive for prevalent CIN 2 or worse tional studies in developing countries (table 2) [26]. The
lesions, but significantly less specific. Whether this gain wide range in accuracy parameters of VIA in different
represents overdiagnosis or protection against future studies underscores the subjective nature of the test, the
high-grade CIN or cervical cancer is not clear. Reduced varying competency of test providers, and the varying
incidence of or mortality from invasive cervical cancer quality of reference standards used to establish the true
among HPV-screened subjects compared to cytologically positive disease. When conventional cytology was con-
screened subjects has not yet been demonstrated and a currently evaluated, the sensitivity of VIA was found to
cluster-randomized trial is addressing this issue in India be higher than or similar to that of cytology, but had low-
[38]. Interim results from this trial showed similar detec- er specificity [26]. It appears that VIA has an average sen-
tion rates of CIN 2 and 3 lesions per 1,000 screened wom- sitivity around 55% and specificity around 85% to detect
en among those screened by cytology or HPV testing [38]. high-grade CIN in experimental study settings.
The reason for this lack of difference in detection rate is The immediate availability of test results following vi-
not obvious at present. HPV testing, reportedly, does not sual testing has opened up the option of ‘screen and treat’
add significant psychological distress when combined or ‘single visit’ approach to ensure a high compliance to
with cytology in routine primary cervical screening [39]. treatment of screen-positive women. In this, screen-pos-
itive women with no clinical evidence of invasive cancer
Visual Screening and satisfying the criteria for ablative therapy are imme-
Naked eye visibility of most precancerous and early diately treated with cryotherapy, without confirmatory
cancerous lesions after application of dilute acetic acid or investigations such as colposcopy or histology. The safe-
Lugol’s iodine solution and the need for affordable, sim- ty, acceptability, and the feasibility of combining VIA
ple cervical screening tests have prompted the evaluation and cryotherapy in a single-visit approach have been
of visual screening tests in comparison with convention- demonstrated in rural Thailand [44], Ghana [45], Guate-
al cytology in recent years. VIA, also known as direct mala [46] and South Africa [47]. VIA-based screen and
visual inspection (DVI), or as acetic acid test (AAT), or treat programmes are currently operational in 15 prov-
cervicoscopy, is the most widely evaluated visual screen- inces in Thailand. In a randomized controlled trial in
ing test. South Africa, VIA followed by cryotherapy resulted in a
VIA involves naked eye inspection of the cervix using 37 and 46% lower prevalence of CIN 2–3 lesions at 6 and
a bright torch light or a halogen focus lamp, 1–2 min after 12 months follow-up compared with a control group [47].
the application of 3–5% acetic acid using a cotton swab or In this study, cryotherapy for HPV test-positive women
a spray. A positive test is characterized by well-defined resulted in much higher decline in the prevalence of CIN
acetowhite areas close to the squamocolumnar junction 2–3 at 6 and 12 months (77 and 74%, respectively). It was
or to the external os or on the entire cervix or a cervical concluded that both screen-and-treat approaches are safe
growth turning acetowhite [40]. Immediate results fol- and result in a lower prevalence of high-grade cervical

Screening and Therapy of Cervical Med Princ Pract 2008;17:351–364 357

Cancer
cancer precursor lesions compared with delayed evalua- cervical growth. The sensitivity of VILI varied between
tion at both 6 and 12 months. 44 and 92% and specificity between 75 and 85% in cross-
Currently, the efficacy and effectiveness of VIA screen- sectional studies (table 2) [26, 50–52].
ing in reducing cervical cancer incidence and mortality
are being addressed in randomized controlled trials in Cost-Effectiveness of New Paradigms of VIA/HPV
India [38, 48]. In a cluster-randomized controlled trial, of Screening
the 114 study clusters in Dindigul district, 57 were ran- Cost effectiveness studies based on data from India,
domized to a single round of VIA by trained nurses and Kenya, Peru, South Africa, and Thailand indicate that the
57 to a control group [48]. Of the 49,311 eligible women most cost-effective strategies for cervical screening are
aged 30–59 years in the VIA group, 31,343 (63.6%) were those approaches requiring the fewest visits, leading
screened; 30,958 women in the control group received to improved follow-up testing and treatment [53, 54].
routine care. During the 7 years following the beginning Screening women once in their lifetime, at the age of
of screening, there were 167 cervical cancer cases and 83 35 years, with a one- or two-visit screening strategy in-
cervical cancer deaths in the intervention group, com- volving VIA or HPV testing in cervical cell samples re-
pared with 158 cases and 92 deaths and in the control duced the lifetime risk of cancer by approximately 25–
group [incidence hazard ratio 0.75 (95% CI 0.55–0.95) 36%, and cost less than 500 dollars per year of life saved
and mortality hazard ratio 0.65 (95% CI 0.47–0.89)] [48]. [54]. Relative cancer risk declined by an additional 40%
These results show a 25% reduction in cancer incidence with two screenings (at 35 and 40 years of age), resulting
and a 35% reduction in cancer mortality. The greatest re- in a cost per year of life saved that was less than each
duction in incidence and mortality rates were observed country’s per capita gross domestic product, a very cost-
for the 30- to 39-year age group, which makes biological effective result [54].
sense, since the transformation zone where cervical neo-
plasia occurs is fully exposed on the ectocervix in young
women, enabling VIA to detect the abnormalities. The Treatment of Cervical Cancer Precursor Lesions
convincing reduction in disease burden and the feasibil-
ity justify the use of VIA screening both in clinical and Principles of Treatment of CIN
public health settings in developing countries. However, The treatment of CIN has evolved from in-patient pro-
good training of providers and sustained quality assur- cedures like hysterectomy and cold knife conization to-
ance are vital for VIA screening to succeed in preventing wards more conservative, yet safe and effective approach-
cervical cancer in developing countries. It is possible that es such as destruction (ablation) or removal (excision) of
the protection from a single screening may diminish with the entire transformation zone of the cervix including the
time and further follow-up in this study will document extension into the crypts (average depth 5 mm). Ablative
the evolution of invasive disease incidence over time. therapy can be done by cryotherapy, carbon dioxide laser
However, the protection observed 6 years from the begin- or cold coagulation. Excisional techniques include large
ning of VIA screening in this study is important from a loop excision of the transformation zone (LLETZ), also
public health point of view for low resource settings. VIA known as LEEP, needle excision of the transformation
may also be a suitable screening approach for the under- zone (NETZ) and laser excision, cold knife conization
served socio-economically disadvantaged populations and hysterectomy. Currently, cold knife conization under
who are currently not covered by the existing cytology local or general anaesthesia is reserved for the treatment
screening programmes in developed countries. of micro-invasive cancer where evaluation of the margin
Studies evaluating low-level magnification for visual- is of prime importance. For adenocarcinoma in situ and
ization of acetowhite changes have shown that it does not previous failed treatment, NETZ may be a better choice.
improve accuracy over that of naked eye visualization Hysterectomy should be reserved only for a select few
[26, 49]. VILI involves naked eye examination of the cer- cases of CIN coexisting with associated gynaecological
vix, to identify mustard-yellow lesions in the transforma- conditions requiring removal of the uterus.
tion zone of the cervix, after application of Lugol’s iodine CIN 2 and CIN 3, being true cervical cancer precur-
[40]. The VILI test results are reported immediately after sors, should always be treated. CIN 1 lesions should be
application of iodine. A positive result is based on the ap- treated if follow-up cannot be ensured (as in most low-
pearance of definite mustard-yellow area on the cervix resource settings) or the lesion persists for 2 years or
close to the squamocolumnar junction or the os or on a worsens in grade or size [1]. In the case of CIN associated

358 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu

with pregnancy, the woman is reassessed 6 months after heat from a high-voltage electrical arc between the oper-
delivery to allow the uterus and cervix to involute and ating electrode and tissue allows the operator to cut by
treatment is decided based on clinical findings at that vaporizing the tissue. Once the specimen has been re-
time. We limit our discussion of treatment techniques to moved and placed in formalin, any bleeding points are
cryotherapy and LLETZ [55]. carefully fulgurated using a ball electrode. Extensive ab-
lation of the treatment crater should be avoided because
Cryotherapy this may interfere with the early detection of inadequate-
Cryotherapy using nitrous oxide or carbon dioxide re- ly treated occult invasive disease. The excision of trans-
frigerant is the simplest, safest and most widely practised formation zone treats the abnormality effectively and
ablative treatment that can be carried out by a wide range provides a specimen for detailed histological evaluation.
of health care personnel like nurses, general practitioners The width of the loops ranges from 10 to 20 mm and the
and specialists. Any grade of CIN can be treated by cryo- depth ranges from 8 to 15 mm. The appropriate size of
therapy provided that the entire lesion is visible and oc- the loop is chosen to achieve adequate depth and width
cupies less than three fourths of the transformation zone; of cut depending on the size and position of the lesion.
it does not extend into endocervix or vagina; no suspicion Though infrequent, excessive bleeding may occur
of glandular or invasive disease and the entire lesion can during or immediately after surgery. Usually such bleed-
be adequately covered by the cryoprobe [55]. Adequate ing can be controlled by diathermy fulguration or by ap-
punch biopsies should be obtained from the abnormal plying Monsel’s paste. Rarely, lateral suturing may be re-
area prior to ablative therapy. Freezing should be done in quired. No significant differences in the frequencies of
two cycles of 3 min with 5 min of thawing in between primary or secondary haemorrhage following LLETZ
(double-freeze technique) [55]. Cryotherapy is a safe pro- and laser ablation have been observed in randomized tri-
cedure with no significant operative morbidity. Cure als [60–62]. Up to 20% of the post-LLETZ biopsy speci-
rates following cryotherapy have reported to vary be- mens may have disease at the margin and such women
tween 86 and 95% for all grades of CIN, 91 and 100% for have a high risk of post-treatment recurrence [63, 64].
CIN 1, 75 and 96% for CIN 2 and 70 and 92% for CIN 3 While every effort should be taken to avoid incomplete
lesions [56, 57]. This impressive success rate of cryother- excision, some of the cases with deep positive margins
apy, like any other excision treatment for CIN, may fall would be safer with a second treatment, but most will
over a long period. Pooled analysis of studies comparing need close follow-up [63, 64]. The risk of recurrence is
laser ablation and cryotherapy failed to demonstrate any higher if the endocervical margin has residual disease.
significant difference in the frequency of residual disease The failure rate of LLETZ varies between 4 and 18% in
(odds ratio: 0.96; 95% CI: 0.67–1.36) [58] and there was no various studies [58, 62, 64]. High-grade post-treatment
statistically significant difference in cure rates of ectocer- disease occurred in 597 of 3,335 (18%) women who had
vical CIN lesions following cryotherapy, laser ablation, incomplete excision versus 318 of 12,493 (3%) women
LEEP and cone biopsy [59]. There is no reason to suspect with complete excision in a recent meta-analysis [64]. The
that high cure rates can be achieved for ectocervical le- randomized trials that compared residual disease at fol-
sions occupying less than three fourths of the transfor- low-up did not observe any significant difference be-
mation zone by cryotherapy. Cryotherapy is an impor- tween LLETZ, laser conization or knife conization [59].
tant and effective work horse for treating ectocervical In a randomized controlled trial comparing NETZ with
CIN in the low- and medium-resource countries, where LEEP involving 347 women with CIN, more women in
most women do not have the means or luxury to access the NETZ arm had clear histological margins (85 vs. 75%)
excisional methods of treatment. Realities are very differ- and produced more specimens in one piece (97.5 vs.
ent in different parts of the world. 29.5%) [65].

Large Loop Excision of the Transformation Zone

(LLETZ or LEEP) Early Detection and Treatment of Invasive Cervical
All CIN lesions including the glandular abnormalities Cancer
can be treated by LLETZ. A wire loop electrode powered
by an electrosurgical unit is used to remove the entire Public and professional awareness are important in
transformation zone, containing the entire lesion, under the early detection and management of invasive cervical
colposcopic control and under local anaesthesia. The cancer. Ensuring availability and accessibility to basic di-

Screening and Therapy of Cervical Med Princ Pract 2008;17:351–364 359

Cancer
Table 3. Treatment methods for cervical cancer

Stage Description of the stage of disease Treatment modality 5-year survival, %

IA invasive cancer identified only microscopically; invasion is stage IA1: cold knife conization or simple hysterectomy 90–95
limited to measured stromal invasion with a maximum stage IA2: radical hysterectomy and bilateral pelvic
depth of 5 mm and no wider than 7 mm lymphadenectomy
– stage IA1: measured invasion of the stroma 3 mm or less or
in depth and 7 mm or less in diameter radical trachelectomy for fertility preservation plus bilateral
– stage IA2: measured invasion of stroma more than 3 mm pelvic lymphadenectomy
but 5 mm or less in depth and 7 mm or less in diameter or
intracavitary radiotherapy, 75–80 Gy to point A
IB clinical lesions confined to the cervix or preclinical lesions stage IB1: radical hysterectomy and bilateral pelvic 80–85
greater than stage IA lymphadenectomy
– stage IB1: clinical lesions 4 cm or less in size or
– stage IB2: clinical lesions more than 4 cm in size radical radiotherapy (external radiotherapy plus
intracavitary radiation)
stage IB2: radical radiotherapy to a total dose of 80–85 Gy to
point A (external radiotherapy plus intracavitary radiation)
with or without concurrent chemotherapy cisplatin or
cisplatin/fluorouracil
or
radical hysterectomy and bilateral pelvic lymphadenectomy
with or without postoperative total pelvic radiation therapy
plus chemotherapy cisplatin or cisplatin/fluorouracil
II stage II is carcinoma that extends beyond the cervix but has stage IIA: radical radiotherapy (external radiotherapy plus 50–65 for IIA
not extended onto the pelvic wall; the carcinoma involves intracavitary radiation) with or without concurrent disease
the vagina but not as far as the lower third section chemotherapy cisplatin or cisplatin/fluorouracil
– stage IIA: no obvious parametrial involvement; or 40–50 for IIB
involvement of as much as the upper two thirds of the radical hysterectomy and bilateral pelvic lymphadenectomy disease
vagina with or without postoperative total pelvic radiation therapy
– stage IIB: obvious parametrial involvement but not onto plus chemotherapy
the pelvic sidewall stage IIB: radical radiotherapy (external radiotherapy plus
intracavitary radiation) with or without concurrent
chemotherapy with cisplatin or cisplatin/fluorouracil
III carcinoma that has extended onto the pelvic sidewall and/or radical radiotherapy (external radiotherapy plus 25–30
involves the lower third of the vagina; on rectal examination, intracavitary radiation) with or without concurrent
there is no cancer-free space between the tumour and the chemotherapy with cisplatin or cisplatin/fluorouracil
pelvic sidewall; all cases with a hydronephrosis or non-
functioning kidney are stage IIIB, unless they are known to
be due to other causes
– stage IIIA: no extension onto the pelvic sidewall but
involvement of the lower third of the vagina
– stage IIIB: extension onto the pelvic sidewall or
hydronephrosis or non-functioning kidney
IV carcinoma that has extended beyond the true pelvis or has stage IVA: radical radiotherapy (external radiotherapy plus <5
clinically involved the mucosa of the bladder and/or rectum intracavitary radiation) with or without concurrent
– stage IVA: spread of the tumour onto adjacent pelvic chemotherapy with cisplatin or cisplatin/fluorouracil
organs such bladder or rectum or
– stage IVB: spread to distant organs palliative radiotherapy
stage IVB: palliative radiotherapy or chemotherapy

agnosis and treatment services is key to the success of of invasive cancer should prompt physical examination
treatment of invasive cervical cancer. and investigations to rule out cancer. Clinical suspicion
Early, preclinical invasive cervical cancers may be de- and speculum examination are important in the early de-
tected during colposcopic assessment of screen-positive tection of invasive cancer. Once a diagnosis of invasive
women. As invasion progresses, symptoms manifest with cancer is made, it is mandatory to stage the clinical extent
characteristic clinical features, depending on the clinical of disease, according to the International Federation of
spread of the disease. Awareness of symptoms and signs Gynaecology and Obstetrics (FIGO) classification, to

360 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu

guide treatment and prognosis. The FIGO clinical stag- The most frequently used surgical procedure for stages
ing is based on primary tumour size and the extent of the IB1 and IIA cervical cancer is Wertheim’s operation (rad-
spread of disease in the vagina, parametrium and distant ical abdominal hysterectomy with bilateral pelvic lymph-
organs (table 3). Speculum, vaginal and rectal examina- adenectomy) with some modifications. The modifica-
tion, chest X-ray and intravenous pyelography or abdom- tions are: the cardinal and the uterosacral ligaments are
inal ultrasound scanning are mandatory investigations removed short of their distal extensions instead of dis-
for staging. In choosing optimal therapy, the clinical secting them right up to their attachments to pelvic walls;
stage of cancer is probably the most fruitful, if not the the ureters are mobilized less, so as to keep the blood sup-
most perfect, surrogate for tumour volume that can also ply to the terminal part intact. Postoperative pelvic irra-
be measured universally. Surgery and radiotherapy, with diation is given to patients with lymph node metastasis,
or without chemotherapy, are important treatment mo- positive vaginal margin, deep stromal invasion or exten-
dalities for invasive cervical cancer (table 3). sive lymphovascular space involvement.
Alternatively stage IB1 cancers may be treated with ex-
Management of Micro-Invasive Cancer ternal-beam pelvic radiation therapy combined with in-
The management of micro-invasive cancer depends tracavitary brachytherapy. Radical radiotherapy compris-
on the depth of stromal invasion. The frequency of nodal es of external beam radiotherapy to the pelvis, to a total
metastases increases from !1% in stage IA1 (stromal in- dose of 45–48 Gray (Gy) in 4 weeks and intracavitary ra-
vasion !3 mm) to 7% in stage IA2 (stromal invasion 3.1– diation 30–35 Gy to point A. Although low-dose rate
5.0 mm) [64]. Stage IA1 disease may be treated by cold brachytherapy, typically with cesium-137, has been the
knife conization alone or simple hysterectomy (table 3). traditional approach, the use of high-dose rate therapy,
If the margins of the cone are free of disease and the depth typically with iridium-192, is rapidly increasing. High-
of invasion does not exceed 3 mm, no further treatment dose rate brachytherapy provides the advantage of elimi-
is necessary. However, conservative management with nating radiation exposure to medical personnel, a shorter
conization is not recommended in this stage in develop- treatment time, patient convenience and outpatient man-
ing countries, due to difficulties in ensuring long-term agement. High-dose rate brachytherapy is comparable
follow-up and adequate histological examination of the with low-dose rate brachytherapy in terms of local-region-
resected specimen. Hysterectomy should be done if the al control and complication rates [68, 69]. It is emphasized
margins are involved. Extrafascial hysterectomy is an op- that the addition of intracavitary irradiation to external
tion to treat stage IA1 disease in women above 40 years beam radiotherapy is associated with improved disease
who have completed child-bearing. Stage IA2 carcinoma control and survival, as compared to external radiothera-
is treated by radical hysterectomy and bilateral pelvic py alone. Stage IB2 tumours have higher chance of having
lymphadenectomy (table 3). Radical trachelectomy in- pelvic and para-aortic lymph node metastasis and are con-
volving pelvic laparoscopic lymphadenectomy followed sidered unsuitable for surgery. Radiotherapy plus chemo-
by vaginal resection of cervix, parametrium and upper therapy with cisplatin or cisplatin/5-fluorouracil is the
one third of the vagina may be performed in selected pa- treatment of choice for stage IB2 disease [68].
tients desirous of preserving fertility. To prevent prema-
ture delivery a prophylactic cerclage is placed at the isth- Management of Stage II–IV Cancer
mus. In medically unfit patients, brachytherapy alone Radiotherapy plus concurrent chemotherapy with cis-
providing a total dose of 75–80 Gy to point A is an equal- platin or cisplatin plus fluorouracil has emerged as the
ly effective alternative treatment option. treatment of choice for those with stages IIB and III cer-
vical cancers [70–73]. Randomized clinical trials have
Management of Stage IB and IIA Cancer confirmed that the combination of chemotherapy and ra-
There is no significant difference in outcome of pa- diotherapy is superior to pelvic radiation alone [68–71].
tients with stage IB1 and IIA cancer treated by irradiation Concurrent chemotherapy may improve the outcome
alone or by radical hysterectomy. The cure rate for either only if radical radiotherapy is fully and accurately ad-
modality of treatment is around 85% at these stages [67]. ministered.
However, surgery is the treatment of choice because of the The introduction of three-dimensional conformal ra-
following advantages: preservation of ovarian function, diotherapy and intensity-modulated radiotherapy have
vaginal pliability, function, and the availability of radia- resulted in overall reduction in volume of normal tissue
tion as a reserved treatment option in case of recurrence. irradiated which may translate into overall reduction in

Screening and Therapy of Cervical Med Princ Pract 2008;17:351–364 361

Cancer
both acute and late treatment-related side effects. Further cancer. Both screening and vaccination have the poten-
trials are necessary to evaluate this perspective and pos- tial to save many lives. At the public health level, health
sible improved cure rates [74, 75]. care infrastructure, affordability and capacity for initiat-
Patients with stage IVA, who already have bladder ing and sustaining vaccination and screening pro-
and/or rectal involvement, may be treated with radical grammes are critical factors in cervical cancer control.
radiotherapy as for stage III patients. However, the pos- Whereas the individual practitioner can successfully use
sibility of urinary or faecal incontinence or vesicovaginal the opportunities provided by health care interactions of
or rectovaginal fistula may be accentuated when the can- women to offer one or more of the several useful inter-
cer regresses due to treatment, necessitating urethrosto- ventions to prevent invasive cancer or death from inva-
my or colostomy. sive cancer by early clinical diagnosis, prompt referral
Stage IVB cervical cancer is incurable with locore- and ensuring appropriate and effective treatment are
gional treatment such as radiotherapy due to involvement necessary. Such efforts by informed clinicians are critical
of distant organs or systemic treatment with chemother- particularly in countries with no organized public health
apy. They are candidates for palliative radiotherapy and/ programme of screening and early detection.
or chemotherapy to control bleeding, excessive discharge
and pain and for symptomatic management.
Acknowledgements

The authors gratefully acknowledge Dr. Kunnambath Rama-

Conclusions
das, Visiting Scientist, Screening Group, International Agency for
Research on Cancer, for his valuable comments on a draft copy
The long natural history of cervical cancer presents of this article. They are also grateful to Mrs. Evelyn Bayle and
several opportunities in terms of prevention, screening, Mrs. Krittika Guinot for their assistance in preparing the manu-
early detection and treatment of CIN to prevent invasive script.

References

1 Ferlay J, Bray F, Pisani P, Parkin DM: Cancer 7 Wang H, Chia KS, Du WB, Lee J, Sankarana- 13 Brown DR, Fife KH, Wheeler CM, Koutsky
Incidence, Mortality and Prevalence World- rayanan R, Sankila R, Sng I, Seow A, Lee HP: LA, Lupinacci LM, Railkar R, Suhr G, Barr
wide. GLOBOCAN 2002. IARC Cancer Base Population-based survival for cervical can- E, Dicello A, Li W, Smith JF, Tadesse A, Jan-
5, ed 2. Lyon, IARC Press, 2004. cer in Singapore, 1968–1992. Am J Obstet sen KU: Early assessment of the efficacy of a
2 Muñoz N, Bosch FX, de Sanjosé S, Herrero Gynecol 2003; 188:324–329. human papillomavirus type 16 L1 virus-like
R, Castellsagué X, Shah KV, Snijders PJ, Mei- 8 Ries LAG, Eisner MP, Kosary CL, et al (eds): particle vaccine. Vaccine 2004; 22: 2936–
jer CJ: International Agency for Research on SEER Cancer Statistics Review, 1975–2001. 2942.
Cancer Multicenter Cervical Cancer Study Bethesda, National Cancer Institute, 2004. 14 Harper DM, Franco EL, Wheeler C, Ferris
Group: Epidemiologic classification of hu- Available at: http://seer.cancer.gov/csr/1975_ DG, Jenkins D, Schuind A, Zahaf T, Innis B,
man papillomavirus types associated with 2001/. Naud P, De Carvalho NS, Roteli-Martins
cervical cancer. N Engl J Med 2003;348:518– 9 The EUROCARE Working Group. EURO- CM, Teixeira J, Blatter MM, Korn AP, Quint
527. CARE-3: survival of cancer patients diag- W, Dubin G; GlaxoSmithKline HPV Vaccine
3 Parkin DM, Whelan SL, Ferlay J, Teppo L, nosed 1990–94 – results and commentary. Study Group: Efficacy of a bivalent L1 virus-
Thomas DB: Cancer Incidence in Five Con- Ann Oncol 2003;14:v61–v118. like particle vaccine in prevention of infec-
tinents. IARC Sci Publ No 155. Lyon, IARC 10 Muñoz N, Bosch FX, Castellsagué X, Díaz M, tion with human papillomavirus types 16
Press, 2002, vol VIII. de Sanjose S, Hammouda D, Shah KV, Meijer and 18 in young women: a randomised con-
4 Gondos A, Brenner H, Wabinga H, Parkin CJ: Against which human papillomavirus trolled trial. Lancet 2004;364:1757–1765.
DM: Cancer survival in Kampala, Uganda. types shall we vaccinate and screen? The in- 15 Harper DM, Franco EL, Wheeler CM,
Br J Cancer 2005;92:1808–1812. ternational perspective. Int J Cancer 2004; Moscicki AB, Romanowski B, Roteli-Mar-
5 Chokunonga E, Ramanakumar AV, Nyaka- 111:278–285. tins CM, Jenkins D, Schuind A, Costa
bau AM, Borok MZ, Chirenje ZM, Sankila R, 11 Smith JS, Lindsay L, Hoots B, Keys J, Fran- Clemens SA, Dubin G; HPV Vaccine Study
Parkin DM: Survival of cervix cancer pa- ceschi S, Winer R, Clifford GM: Human Group: Sustained efficacy up to 4.5 years of
tients in Harare, Zimbabwe, 1995–1997. Int J papillomavirus type distribution in invasive a bivalent L1 virus-like particle vaccine
Cancer 2004;109:274–277. cervical cancer and high-grade cervical le- against human papillomavirus types 16 and
6 Sankaranarayanan R, Black RJ, Parkin DM: sions: a meta-analysis update. Int J Cancer 18: follow-up from a randomised control tri-
Cancer Survival in Developing Countries. 2007;121:621–632. al. Lancet 2006;367:1247–1255.
IARC Sci Publ No 15. Lyon, IARC Press, 12 Clifford GM, Smith JS, Plummer M, Muñoz
1988. N, Franceschi S: Human papillomavirus
types in invasive cancer worldwide: a meta-
analysis. Br J Cancer 2003;88:63–73.

362 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu

16 Koutsky LA, Ault KA, Wheeler CM, Brown 26 Sankaranarayanan R, Gaffikin L, Jacob M, 37 Stewart DE, Gagliardi A, Johnston M,
DR, Barr E, Alvarez FB, Chiacchierini LM, Sellors J, Robles S: A critical assessment of Howlett R, Barata P, Lewis N, Oliver T, Mai
Jansen KU; Proof of Principle Study Investi- screening methods for cervical neoplasia. Int V: Self-collected samples for testing of onco-
gators: A controlled trial of a human papil- J Gynaecol Obstet 2005;89(suppl 2):S4–S12. genic human papillomavirus: a systematic
lomavirus type 16 vaccine. N Engl J Med 27 IARC Handbooks on Cancer Prevention. review. J Obstet Gynaecol Can 2007; 29:817–
2002;347:1645–1651. Lyon, IARC Press, 2004, vol 10: Cervix Can- 828.
17 Mao C, Koutsky LA, Ault KA, Wheeler CM, cer Screening. 38 Sankaranarayanan R, Nene BM, Dinshaw
Brown DR, Wiley DJ, Alvarez FB, Bautista 28 Sepulveda CC, Prado R: Effective cervical KA, Mahe C, Jayant K, Shastri SS, Malvi SG,
OM, Jansen KU, Barr E: Efficacy of human cytology screening programmes in middle- Chinoy R, Kelkar R, Budukh AM, Keskar V,
papillomavirus-16 vaccine to prevent cervi- income countries: the Chilean experience. Rajeshwarker R, Muwonge R, Kane S, Parkin
cal intraepithelial neoplasia: a randomized Cancer Detect Prev 2005;29:405–411. DM, Chauhan MK, Desai S, Fontaniere B,
controlled trial. Obstet Gynecol 2006; 107: 29 Klinkamer PJ, Meerding WJ, Rosier PF, Frappart L, Kothari A, Lucas E, Panse N; Os-
18–27. Hanselaar AG: Liquid-based cervical cytol- manabad District Cervical Screening Study
18 Villa LL, Costa RL, Petta CA, Andrade RP, ogy. Cancer 2003;99:263–271. Group: A cluster randomized controlled tri-
Ault KA, Giuliano AR, Wheeler CM, Kout- 30 Davey E, Barratt A, Irwig L, Chan SF, Ma- al of visual, cytology and human papilloma-
sky LA, Malm C, Lehtinen M, Skjeldestad caskill P, Mannes P, Saville AM: Effect of virus screening for cancer of the cervix in
FE, Olsson SE, Steinwall M, Brown DR, Kur- study design and quality on unsatisfactory rural India. Int J Cancer 2005;116:617–623.
man RJ, Ronnett BM, Stoler MH, Ferenczy A, rates, cytology classifications, and accuracy 39 Kitchener HC, Fletcher I, Roberts C, Wheel-
Harper DM, Tamms GM, Yu J, Lupinacci L, in liquid-based versus conventional cervical er P, Almonte M, Maguire P: The psychoso-
Railkar R, Taddeo FJ, Jansen KU, Esser MT, cytology: a systematic review. Lancet 2006; cial impact of human papillomavirus testing
Sings HL, Saah AJ, Barr E: Prophylactic 367:122–132. in primary cervical screening: a study within
quadrivalent human papillomavirus (types 31 Ronco G, Cuzick J, Pierotti P, Cariaggi MP, a randomized trial. Int J Gynecol Cancer
6, 11, 16, and 18) L1 virus-like particle vac- Dalla Palma P, Naldoni C, Ghiringhello B, 2007, E-pub ahead of print.
cine in young women: a randomised double- Giorgi-Rossi P, Minucci D, Parisio F, Pojer A, 40 Sankaranarayanan R, Wesley R: A Practical
blind placebo-controlled multicentre phase Schiboni ML, Sintoni C, Zorzi M, Segnan N, Manual on Visual Screening for Cervical
II efficacy trial. Lancet Oncol 2005; 6: 271– Confortini M: Accuracy of liquid based ver- Neoplasia. IARC Tech Publ No 41. Lyon,
278. sus conventional cytology: overall results of IARC Press, 2003.
19 Villa LL, Ault KA, Giuliano AR, Costa RL, new technologies for cervical cancer screen- 41 Blumenthal PD, Lauterbach M, Sellors JW,
Petta CA, Andrade RP, Brown DR, Ferenczy ing: randomised controlled trial. BMJ 2007; Sankaranarayanan R: Training for cervical
A, Harper DM, Koutsky LA, Kurman RJ, 335:28–31. cancer prevention programs in low-resource
Lehtinen M, Malm C, Olsson SE, Ronnett 32 Franco EL: Primary screening of cervical settings: focus on visual inspection with ace-
BM, Skjeldestad FE, Steinwall M, Stoler MH, cancer with human papillomavirus tests. J tic acid and cryotherapy. Int J Gynaecol Ob-
Wheeler CM, Taddeo FJ, Yu J, Lupinacci L, Natl Cancer Inst Monogr 2003;31:89–96. stet 2005;89(suppl 2):S30–S37.
Railkar R, Marchese R, Esser MT, Bryan J, 33 Koliopoulos G, Arbyn M, Martin-Hirsch P, 42 Mcintosh N, Blumenthal PD, Blouse A (eds):
Jansen KU, Sings HL, Tamms GM, Saah AJ, Kyrgiou M, Prendiville W, Paraskevaidis E: Cervical Cancer Prevention Guidelines for
Barr E: Immunologic responses following Diagnostic accuracy of human papillomavi- Low-Resource Settings. JHPIEGO Technical
administration of a vaccine targeting human rus testing in primary cervical screening: a Manual. Baltimore, JHPIEGO, 2001.
papillomavirus types 6, 11, 16, and 18. Vac- systematic review and meta-analysis of non- 43 http://screening.iarc.fr
cine 2006;24:5571–5583. randomized studies. Gynecol Oncol 2007; 44 Gaffikin L, Blumenthal PD, Emerson M,
20 Koutsky LA, Harper DM: Chapter 13. Cur- 104:232–246. Limpaphayom K: Safety, acceptability, and
rent findings from prophylactic HPV vac- 34 Naucler P, Ryd W, Törnberg S, Strand A, feasibility of a single-visit approach to cer-
cine trials. Vaccine 2006; 24(suppl 3):S114– Wadell G, Elfgren K, Rådberg T, Strander B, vical-cancer prevention in rural Thailand:
S121. Forslund O, Hansson BG, Rylander E, Dill- a demonstration project. Lancet 2003; 361:
21 The Future II Study Group: Quadrivalent ner J: Human papillomavirus and Papanico- 814–820.
vaccine against human papillomavirus to laou tests to screen for cervical cancer. N 45 Blumenthal PD, Gaffikin L, Deganus S, Lew-
prevent high-grade cervical lesions. N Engl J Engl J Med 2007;357:1589–1597. is R, Emerson M, Adadevoh S: Cervical can-
Med 2007;356:1915–1927. 35 Mayrand MH, Duarte-Franco E, Rodrigues cer prevention: safety, acceptability, and fea-
22 Agosti J, Goldie S: Introducing HPV vaccine I, Walter SD, Hanley J, Ferenczy A, Ratnam sibility of a single-visit approach in Accra,
in developing countries: key challenges and S, Coutlée F, Franco EL; Canadian Cervical Ghana. Am J Obstet Gynecol 2007;196:407–
issues. N Engl J Med 2007;356:1908–1910. Cancer Screening Trial Study Group: Hu- 408.
23 Sankaranarayanan R, Budukh A, Rajkumar man papillomavirus DNA versus Papanico- 46 Mathers LJ, Wigton TR, Leonhardt JG:
R: Effective screening programs for cervical laou screening tests for cervical cancer. N Screening for cervical neoplasia in an un-
cancer in low- and middle-income develop- Engl J Med 2007;357:1579–1588. selected rural Guatemalan population using
ing countries. Bull World Health Organ 36 Bulkmans N, Berkhof J, Rozendaal L, van direct visual inspection after acetic acid ap-
2001;79:954–962. Kemenade F, Boeke A, Bulk S, Voorhorst F, plication: a pilot study. J Low Genit Tract Dis
24 Fahey MT, Irwig L, Macaskill P: Meta-analy- Verheijen R, van Groningen K, Boon M, 2005;9:232–235.
sis of Pap test accuracy. Am J Epidemiol Ruitinga W, van Ballegooijen M, Snijders P, 47 Denny L, Kuhn L, De SM, Pollack AE, Du-
1995;141:680–689. Meijer C: Human papillomavirus DNA test- pree W, Wright TC Jr: Screen-and-treat ap-
25 Nanda K, McCrory DC, Myers ER, Bastian ing for the detection of cervical intraepithe- proaches for cervical cancer prevention in
LA, Hasselblad V, Hickey JD, Matchar DB: lial neoplasia grade 3 and cancer: 5-year low-resource settings: a randomized con-
Accuracy of the Papanicolaou test in screen- follow-up of a randomised controlled imple- trolled trial. JAMA 2005;294:173–181.
ing for and follow-up of cervical cytologic mentation trial. Lancet 2007; 370: 1764– 48 Sankaranarayanan R, Esmy PO, Rajkumar R,
abnormalities: a systematic review. Ann In- 1772. Muwonge R, Swaminathan R, Shanthaku-
tern Med 2000;132:810–819. mari S, Fayette JM, Cherian J: Effect of visual
screening on cervical cancer incidence and
mortality in Tamil Nadu, India: a cluster-
randomised trial. Lancet 2007; 370:398–406.

Screening and Therapy of Cervical Med Princ Pract 2008;17:351–364 363

Cancer
49 Shastri SS, Dinshaw K, Amin G, Goswami S, 57 Sankaranarayanan R, Rajkumar R, Esmy 67 Landoni F, Maneo A, Colombo A, Placa F,
Patil S, Chinoy R, Kane S, Kelkar R, Mu- PO, Fayette JM, Shanthakumary S, Frappart Milani R, Perego P, Favini G, Ferri L, Man-
wonge R, Mahé C, Ajit D, Sankaranarayanan L, Thara S, Cherian J: Effectiveness, safety gioni C: Randomized study of radical sur-
R: Concurrent evaluation of visual, cytolog- and acceptability of ‘see and treat’ with cryo- gery versus radiotherapy for stage IB-IIA
ical and HPV testing as screening methods therapy by nurses in a cervical screening cervical cancer. Lancet 1997;350:535–540.
for the early detection of cervical neoplasia study in India. Br J Cancer 2007; 96: 738– 68 Hareyama M, Sakata K, Oouchi A, Nagakura
in Mumbai, India. Bull World Health Organ 743. H, Shido M, Someya M, Koito K: High-dose-
2005;83:186–194. 58 Martin-Hirsch PL, Paraskevaidis E, Kitch- rate versus low-dose-rate intracavitary ther-
50 Sankaranarayanan R, Basu P, Wesley RS, ener H: Surgery for cervical intraepithelial apy for carcinoma of the uterine cervix: a
Mahe C, Keita N, Mbalawa CC, Sharma R, neoplasia. Cochrane Database Syst Rev 1999; randomized trial. Cancer 2002; 94:117–124.
Dolo A, Shastri SS, Nacoulma M, Nayama M, 3:CD001318. 69 Lertsanguansinchai P, Lertbutsayanukul C,
Somanathan T, Lucas E, Muwonge R, Frap- 59 Nuovo J, Melnikow J, Willan AR, Chan BK: Shotelersuk K, Khorprasert C, Rojpornpra-
part L, Parkin DM; IARC Multicentre Study Treatment outcomes for squamous intraepi- dit P, Chottetanaprasith T, Srisuthep A, Su-
Group on Cervical Cancer Early Detection: thelial lesions. Int J Gynaecol Obstet 2000; riyapee S, Jumpangern C, Tresukosol D,
Accuracy of visual screening for cervical 68:25–33. Charoonsantikul C: Phase III randomized
neoplasia: results from an IARC multicentre 60 Alvarez RD, Helm CW, Edwards RP, Nau- trial comparing LDR and HDR brachyther-
study in India and Africa. Int J Cancer 2004; mann RW, Partridge EE, Shingleton HM, apy in treatment of cervical carcinoma. Int J
110:907–913. McGee JA, Hall JB, Higgins RV, Malone JM Radiat Oncol Biol Phys 2004;59:1424–1431.
51 Sarian LO, Derchain SF, Naud P, Roteli-Mar- Jr: Prospective randomised trial of LLETZ 70 Thomas GM: Improved treatment for cervi-
tins C, Longatto-Filho A, Tatti S, Branca M, versus laser ablation in patients with cervical cal cancer-concurrent chemotherapy and ra-
Erzen M, Serpa-Hammes L, Matos J, Gontijo intra-epithelial neoplasia. Gynecol Oncol diotherapy. N Engl J Med 1999; 340: 1198–
R, Bragança JF, Lima TP, Maeda MY, Lörincz 1994;52:175–179. 1200.
A, Dores GB, Costa S, Syrjänen S, Syrjänen 61 Gunasekera PC, Phipps J, Lewis B: Large 71 Green JA, Kirwan JM, Tierney JF, Symonds
K: Evaluation of visual inspection with ace- loop excision of the transformation zone P, Fresco L, Collingwood M, Williams CJ:
tic acid (VIA), Lugol’s iodine (VILI), cervical (LLETZ) compared to carbon dioxide treat- Survival and recurrence after concomitant
cytology and HPV testing as cervical screen- ment of CIN: a superior mode of treatment. chemotherapy and radiotherapy for cancer
ing tools in Latin America. This report refers Br J Obstet Gynaecol 1990; 97:995–998. of the uterine cervix: a systematic review and
to partial results from the LAMS (Latin 62 Mitchell MF, Tortolero-Luna G, Cook E, meta-analysis. Lancet 2001;358:781–786.
American Screening) study. J Med Screen Whittaker L, Rhodes-Morris H, Silva E: A 72 Whitney CW, Sause W, Bundy BN, Malfeta-
2005;12:142–149. randomized clinical trial of cryotherapy, no JH, Hannigan EV, Fowler WC Jr, Clarke-
52 Sangwa-Lugoma G, Mahmud S, Nasr SH, laser vaporization, and loop electrosurgical Pearson DL, Liao SY: Randomized compari-
Liaras J, Kayembe PK, Tozin RR, Drouin P, excision for treatment of squamous intraepi- son of fluorouracil plus cisplatin versus
Lorincz A, Ferenczy A, Franco EL: Visual in- thelial lesions of the cervix. Obstet Gynecol hydroxyurea as an adjunct to radiation ther-
spection as a cervical cancer screening meth- 1998;92:737–744. apy in stage IIB-IVA carcinoma of the cervix
od in a primary health care setting in Africa. 63 Murdoch JB, Morgan PR, Lopes A, Monaghan with negative para-aortic lymph nodes: a
Int J Cancer 2006;119:1389–1395. JM: Histological incomplete excision of CIN Gynecologic Oncology Group and South-
53 Mandelblatt JS, Lawrence WF, Gaffikin L, after large loop excision of the transforma- west Oncology Group study. J Clin Oncol
Limpahayom KK, Lumbiganon P, Waraka- tion zone (LLETZ) merits careful follow-up, 1999;17:1339–1348.
min S, King J, Yi B, Ringers P, Blumenthal not retreatment. Br J Obstet Gynaecol 1992; 73 Thomas GM: Concurrent chemotherapy and
PD: Costs and benefits of different strategies 99:990–993. radiation for locally advanced cervical can-
to screen for cervical cancer in less-devel- 64 Ghaem-Maghami S, Saji S, Majeed G, Sout- cer: the new standard of care. Semin Radiat
oped countries. J Natl Cancer Inst 2002; 94: ter WP: Incomplete excision of cervical in- Oncol 2000;10:44–50.
1469–1483. traepithelial neoplasia and risk of treatment 74 Portelance L, Chao KS, Grigsby PW, Bennet
54 Goldie S, Gaffikin L, Goldhaber-Fiebert J, failure: a meta-analysis. Lancet Oncol 2007; H, Low D: Intensity-modulated radiation
Gordillo-Tobar A, Levin C, Mahe C, Wright 8:985–993. therapy (IMRT) reduces small bowel, rec-
T: Cost effectiveness of cervical screening in 65 Panoskaltsis T, Ind TE, Perryman K, Dina R, tum, and bladder doses in patients with cer-
five developing countries. N Engl J Med Abrahams Y, Soutter WP: Needle versus vical cancer receiving pelvic and para-aortic
2005;353:2158–2168. loop diathermy excision of the transforma- irradiation. Int J Radiat Biol Phys 2001; 51:
55 Sellors J, Sankaranarayanan R: Colposcopy tion zone for the treatment of cervical in- 261–266.
and Treatment of Cervical Intraepithelial traepithelial neoplasia: a randomised con- 75 Heron DE, Gerszten K, Selvaraj RN, King
Neoplasia: A Beginners’ Manual. Lyon, trolled trial. BJOG 2004;111:748–753. GC, Sonnik D, Gallion H, Comerci J, Ed-
IARC Press, 2003. 66 Piver MS, Rose PG, Freedman MF: Change wards RP, Wu A, Andrade RS, Kalnicki S:
56 Alliance for Cervical Cancer Prevention: Ef- in International Federation of Gynecology Conventional 3D conformal versus intensi-
fectiveness, Safety, and Acceptability of and Obstetrics staging. Am J Obstet Gynecol ty-modulated radiotherapy for the adjuvant
Cryotherapy: A Systematic Literature Re- 1988;158:678–679. treatment of gynecologic malignancies: a
view. Seattle, PATH, 2003. www.path.org/ comparative dosimetric study of dose-vol-
files/RH_cryo_white_paper.pdf. ume histograms small star, filled. Gynecol
Oncol 2003;91:39–45.

364 Med Princ Pract 2008;17:351–364 Sankaranarayanan /Thara /Esmy /Basu