Professional Documents
Culture Documents
75 Bosch1995 PDF
75 Bosch1995 PDF
796 ARTICLES Journal of the National Cancer Institute, Vol. 87, No. 11, June 7, 1995
that 20%-40% of sexually active young women have detectable Patients for whom histologic confirmation of invasive cancer was not available
HPV infection and that prevalence decreases with age. In most in the medical records and for whom the review of the slides did not confirm the
presence of cancer were ultimately excluded from the study.
studies, HPV 16 has been found to be the most prevalent HPV
type in cytologically normal women, women with cervical in- HPV Detection and Typing
traepithelial neoplasia (CIN), and women with cervical cancer.
There is, at least within the United States, some geographic Crude DNA samples were prepared from the tumor specimens for subsequent
HPV DNA detection. Each tumor was thawed slightly, placed in a sterile dish,
and ethnic variation in the HPV types detected {10-13), but less and liberally rinsed with sterile saline to remove blood and debris. When pos-
is known about the international variation of HPV types in cer- sible, presumptive tumor tissue was identified macroscopically and used for
vical cancer. Moreover, it is debatable whether HPV-negative processing. A 2- to 3-mm3 segment was excised from the tumor, minced to a
cervical cancer is a biologic entity that is etiologically unrelated pulp, and then proteinase treated (16). Of each prepared sample, 2.5% was used
in the HPV DNA analysis. All utensils were disposable and were discarded im-
to HPV, an artifact attributable to limitations of current detec-
mediately after each use (single use for each specimen). Standard methods to
tion methods, or the result of loss of HPV DNA during the avoid and monitor for contamination were used throughout the laboratory
evolution of the tumor. analysis (17).
Through an extensive multinational collaboration, we or- A widely used PCR DNA amplification-based method was employed for the
ganized the International Biological Study on Cervical Cancer detection and typing of HPV, and 26 type-specific probes were utilized
(IBSCC) to further our understanding of the relationship be- (16,18,19). Primers for a fragment of p^globin gene served as an internal control
to assess the sufficiency of each specimen for amplification. All specimens were
tween HPV and cervical cancer. We sought to determine prepared and analyzed in one laboratory (M. M. Manos and K. V. Shah), follow-
Journal of the National Cancer Institute, Vol. 87, No. 11, June 7, 1995 ARTICLES 797
region to 56.5 years (±14.3 years) in the southern European HPV 18-related group (HPV 18, 39, 45, 59, and 68) represented
region. The mean number of cases included per country was 27.3%.
47.0 (range, 12-123) (Table 1). Twenty different viruses were detected by hybridization with
The diagnosis of invasive cervical carcinoma was confirmed type-specific probes at least once each in the tumor collection.
in 798 (85.6%) of 932 of the included patients, based on review HPVs 40, 42, 53, 54, and 66 and papillomavirus novel PAP 155
of the slides provided by the participant institutions. For the (20) were not identified in any of the specimens. HPV types 6
remaining 134 patients, the histologic data were extracted from and 11 were each found once in the collection.
the medical records as indicated in the "Patients and Methods" Two different viruses were detected in 36 (4%) of the
section. The distribution of disease was as follows: squamous specimens, and these are counted twice in Tables 2 and 3.
cell carcinoma—881 (94.5%), adenocarcinoma—25 (2.7%), Double infections did not cluster geographically, and there was
adenosquamous—18 (1.9%), and other or unknown—eight no relationship between double infections and histologic type,
(0.9%). Based on the degree of nuclear and architectural abnor- grade of differentiation, or stage of the disease (data not shown).
malities and the proportion of cells showing cytoplasmic dif- Twelve specimens harbored HPVs that could not be typed
ferentiation (squamous and glandular), disease was graded as with our battery of 26 type-specific probes. By restriction frag-
follows: well differentiated—82 (10.8%), moderately differen- ment length polymorphism analysis and DNA sequencing of the
tiated—405 (53.1%), and poorly differentiated—275 (36.1%); amplification products, 10 specimens appeared to be either pre-
this information was missing for the remaining 170 (18.2%) viously identified types for which we did not probe (e.g., HPV
Africa
Algeria 41 12 3 (25.0) 9 (75.0) 4 (33.3) 5(41.7) 0 (0.0) 0 (0.0)
Benin 12 6 1 (16.7) 5(83.3) 3 (50.0) 1 (16.7) 0 (0.0) 0 (0.0)
Guinea 21 18 0 (0.0) 18(100.0) 7 (38.9) 1 (5.6) 0 (0.0) 6 (33.3)
Mali 59 58 6(10.3) 52 (89.7) 20 (34.5) 7(12.1) 4 (6.9) 10(17.2)
Uganda 48 43 3 (7.0) 40 (93.0) 23 (53.5) 7(16.3) 1 (2.3) 2 (4.7)
Tanzania 51 49 6(12.2) 43 (87.8) 22 (44.9) 12 (24.5) 0 (0.0) 5(10.2)
Central and
South America
Argentina 61 57 3 (5.3) 54 (94.7) 34 (59.6) 8(14.0) 3 (5.3) 3 (5.3)
Bolivia 56 49 4 (8.2) 45(91.8) 17 (34.7) 2(4.1) 13 (26.5) 4 (8.2)
Brazil 49 46 6(13.0) 40 (87.0) 24 (52.2) 4(8.7) 2(4.3) 2 (4.3)
Chile 85 80 6(7.5) 74 (92.5) 36 (45.0) 4 (5.0) 7 (8.8) 7 (8.8)
Columbia 42 38 2(5.3) 36(94.7) 20 (52.6) 3 (7.9) 3 (7.9) 2 (5.3)
Cuba 51 45 3(6.7) 42 (93.3) 26 (57.8) 3 (6.7) 2 (4.4) 3 (6.7)
Panama 80 73 5 (6.8) 68 (93.2) 34 (46.6) 11(15.1) 2(2.7) 7 (9.6)
Paraguay 123 117 7 (6.0) 110 (94.0) 64 (54.7) 13(11.1) 3 (2.6) 9 (7.7)
Southeast Asia
Indonesia 52 47 2(4.3) 45 (95.7) 15(31.9) 23 (48.9) 0 (0.0) 4 (8.5)
Philippines 26 24 1 (4.2) 23 (95.8) 11 (45.8) 2 (8.3) 0 (0.0) 2(8.3)
Thailand 27 27 0 (0.0) 27(100.0) 16(59.3) 6(22.2) 1 (3.7) 2(7-4)
North America
Canada 49 46 3 (6.5) 43 (93.5) 27 (58.7) 8(17.4) 2 (4.3) 5 (10.9)
United States 13 11 1(9.1) 10 (90.9) 6 (54.5) 1 (9.1) 1 (9.1) 3 (27.3)
Europe
Germany 17 17 1 (5.9) 16(94.1) 13 (765) 0 (0.0) 0 (0.0) 0 (0.0)
Poland 25 23 0 (0.0) 23 (100.0) 18 (78.3) 4(17.4) 2(8.7) 0 (0.0)
Spain 47 46 3 (6.5) 43 (93.5) 25 (54.3) 3(6.5) 3 (6.5) 2 (4.3)
Total 1035 932 66 a n 866 (92.9) 465 (49.9) 128(13.7) 49 (5.3) 78 (8.4)
798 ARTICLES Journal of the National Cancer Institute, Vol. 87, No. 11, June 7, 1995
Table 2. Prevalence of individual HPV types by geographic region
HPV 16 and
16 related
16 79 (42.5) 255 (50.5) 42 (42.9) 56(65.1)t 33 (57.9) 465 (49.9)
31 5(27) 35 (6.9) 1 (1.0) 5 (5.8) 3 (5.3) 49 (5.3)
33 5(27) 18(3.6) 2 (2.0) 1 (1.2) 0 (0.0) 26 (2.8)
52 4 (2.2) 16(3.2) 2 (2.0) 3 (3.5) 0 (0.0) 25 (2.7)
58 5(2.7) 11 (2.2) 2 (2.0) 1 (1.2) 0 (0.0) 19(2.0)
35 4 (2.2) 10(2.0) 1 (1.0) 1 (1.2) 0 (0.0) 16(1.7)
HPV 18 and
18 related
18 33(17.7) 48 (9.5)§ 31 (31.6)t 7(8.1) 9(15.8) 128(13.7)
45 23 (12.4)4: 37(7.3) 8 (8.2) 2 (2.3) 8(14.0) 78 (8.4)
59 0 (0.0) 14(2.8)t 1 (1.0) 0 (0.0) 0 (0.0) 15(1.6)
39 0 (0.0) 13(2.6)t 1 (1.0) 0 (0.0) 0 (0.0) 14(1.5)
68 4 (2.2) 2 (0.4) 1 (1.0) 3 (3.5) 1 (1.8) 11(1.2)
•HPV types 40, 42, 53, 54, 66 and PAP 155 were not identified. Double infections were counted twice. Samples that were HPV negative were excluded from
statistical analysis.
tSignificant increase (/><.0O5).
^Significant increase (/><.05).
§Significant decrease (V<.005).
Table 3. Prevalence of HPV type by stage of the disease and histologic tumor characteristics4
Histologic type, No. Histologic differentiation degree, No. Clinical stage, No.
(% of total specimens) (% of total specimens) (% of total specimens)
HPV 16 451(51.2) 7 (28.0) 3(16.7) 46(56.1) 199(49.1) 126(45.8) 105 (52.2) 301 (48.0) 16(57.1)
HPV 16 131 (14.9) 0 (0.0) 2(11.1) 9(11.0) 66(16.3) 34(124) 28(13.9) 92(14.7) 3 (10.7)
related
HPV 18 107(12.1) 14(56.0) 7 (38.9) 6(7.3) 59(14.6) 49(17.8) 32(15.9) 86(13.7) 2(7.1)
HPV 18 109(12.4) 3(12.0) 5 (27.8) 8 (9.8) 46(11.4) 47(17.1) 20(10.0) 79(12.6) 5 (17.9)
related
Other 55 (6.2) 0 (0.0) 0 (0.0) 10(122) 24(5.9) 12(4.4) 11(5.5) 39 (6.2) 3(10.7)
HPV negative 64(7.3) 1 (4.0) 1 (5.6) 7(8.5) 26 (6.4) 18(6.5) 13(6.5) 47 (7.5) 2(7.1)
Total HPV 853 24 17 79 394 268 196 597 29
positives
Total No. of 881 (100) 25(100) 18(100) 82(100) 405 (100) 275 (100) 201 (100) 627(100) 28(100)
specimens
x2= 61.2
2 2
/> = .001 X = 203 P = .03 Z = 5.3 P = .87
•Double infections were counted twice. Samples that were HPV negative and for which clinical data were unknown were excluded from statistical tests.
Journal of the National Cancer Institute, Vol. 87, No. 11, June 7,1995 ARTICLES 799
18 and related viruses accounted for 71% of the viral types countries, ranging from 75.0% to 100%. These results are com-
found in adenocarcinomas and 71% in adenosquamous car- patible with the hypothesis that only a small proportion of in-
cinomas. HPV 18 and related types were also associated with vasive cervical cancer cases do not contain HPV DNA. As the
degree of differentiation. Their prevalence showed an increasing techniques to detect viral DNA have improved, the reported
trend with decreasing degree of differentiation, and the associa- prevalence among cervical cancer patients has increased
tion persisted when adjusted for histology. No relationship was dramatically (22). Whether our results are contingent on the
observed between viral type and clinical stage of disease. When HPV detection methods remains to be established. Consistent
the above analyses were repeated including only the cases con- with this idea, analysis of the 66 HPV-negative specimens using
firmed on review, the direction and strength of the associations additional HPV detection methods (e.g., other primers) suggests
remained unchanged. that fewer than 5% of cervical cancers are truly HPV negative.
The distribution of HPV types varied geographically. Table 2 The prevalence of HPV in the general population has been es-
and Fig. 1 show the HPV-specific prevalences by geographic timated in case-control studies conducted in six of the countries
region. HPV 16 was the predominant type in all countries except included in our study. The prevalence in control subjects with
Indonesia, where HPV 18 accounted for more than 50%. This the same age distribution as the cervical cancer patients was 5%
predominance of HPV 18 was not explained by tumor type be- in Spain and between 13% and 20% in Thailand, the Philip-
cause it persisted when the analysis was restricted to squamous pines, Paraguay, Brazil, and Colombia. Prevalence in western
cell tumors. In Mali (western Africa), HPV 45 was in significant Europe and the United States is generally less than 10% at 40
years of age or older (2324).
800 ARTICLES Journal of the National Cancer Institute, Vol. 87, No. 11, June 7,1995
laboratory and by histologic review conducted by one expert equivalent to those posed by infection with the most common
pathologist. In our final analyses, we chose to accept HPV-nega- types (e.g., HPVs 16 and 18, which are termed "high-risk"). Al-
tive results only from specimens with adjacent, confirmed tumor though our understanding of cervical cancer has been en-
tissue in order to avoid false-negative results. Our prevalence lightened by defining the role of HPV, it may be complicated by
estimate (93%) may therefore be slightly inflated because the the intricacies of this diverse group of viruses (6).
only restrictions for HPV-positive specimens were adequate Cervical cancer is the second most common cancer in women
diagnostic confirmation and PCR sufficiency. The overall prev- worldwide. It takes its greatest toll in the developing world,
alence was 87.2% when all 126 HPV-negative specimens were where it is the most common cancer among women and where
included, but this percentage is certainly too low because 36.5% health care resources are limited. The knowledge that a
of these specimens contained no tumor material. heterogeneous family of sexually transmitted agents confers the
Despite suboptimal collection, storage, and transport condi- large majority of risk for this disease has far-reaching implica-
tions in many facilities, a surprising majority (more than 95%) tions for prevention strategies, including the development of ef-
of the specimens yielded material that was sufficiently intact for fective vaccines.
PCR-based HPV detection. During preparation of the specimens
for HPV detection, great care was taken to remove blood,
debris, or obviously necrotic tissue as well as to macroscopical- References
ly identify tumor material. We believe that these procedures (/) Parkin DM, Muir CS, Whelan SL, et al: Cancer incidence in five continents.
Journal of the National Cancer Institute, Vol. 87, No. 11, June 7, 1995 ARTICLES 801
Molecular Diagnostics of Human Cancer (Furth M, Greaves MF, eds). Notes
Cold Spring Harbor, NY: Cold Spring Harbor Press, 1989, pp 209-214
(19) Hildesheim A, Schiffman MH, Gravitt PE, et al: Persistence of type- LBSCC Study Group: E. Alihonou, University Nationale du Benin, Cotonou,
specific human papillomavirus infection among cytologically normal Benin; S. Bayo, Institut National de Recherche en Sant6 Publique, Bamako,
women [see comment citation in Medline]. J Infect Dis 169:235-240, 1994 Mali; H. Cherif Mokhtar, Centre Hospitalo-Universitaire, Setif, Algeria; S.
(20) Manos MM, Waldman J, Zhang TY, et al: Epidemiology and partial Chichareon, Prince of Songkla University, Hat-Yai, Thailand; A. Daudt, Hospi-
nucleotide sequence of four novel genital human papillomaviruses. J Infect tal of the Clinics of Porto Alegre, Brazil; E. de los Rios, National Oncology In-
Dis 170:1096-1099,1994 stitute, Panama; P. Ghadirian, Hfltel-Dieu de Montreal, Montreal, Canada; J. N.
(21) Peyton CL, Jansen AM, Wheeler CM, et al: A novel human papillomavirus Kitinya, Muhimbili Medical Centre, Dar es Salaam, Tanzania; M. Koulibaly,
sequence from an international cervical cancer study. J Infect Dis Centre National d'Anatomie Pathologique, Conakry, Guinea; C. Ngelangel,
170:1093-1095,1994 University of the Philippines, Manila, Philippines; L. M. Puig Tintore, Hospital
(22) Mufioz N, Bosch FX: HPV and cervical neoplasia: a review of case-control Clfnic I Provincial, Barcelona, Spain; J. L. Rios-Dalenz, Cancer Registry of La
and cohort studies. IARC Sci Publ 119:255-266,1992 Paz, La Paz, Bolivia; Sarjadi, Diponegoro University Medical Faculty,
(25) Schiffman MH: Recent progress in defining the epidemiology of human Semarang, Indonesia; A. Schneider, Friedrich Schiller University, Jena, Federal
papillomavirus infection and cervical neoplasia. J Natl Cancer Inst 84:394- Republic of Germany; L. Tafur, University of Valle, Cali, Colombia; A. R.
398, 1992 Teyssie, National Institute of Microbiology, Buenos Aires, Argentina; P. A.
(24) Mufioz N, Bosch FX, Deacon J, et al: HPV and cervical cancen a multi- Rol6n, Registro Nacional de Patologfa Tumoral, Asuncion, Paraguay; M. Tor-
centre case-control study. Abstract presented at the 13th International roella, National Cancer Institute, Havana, Cuba; A. Vila Tapia, Regional Clini-
Papillomavirus Conference, Amsterdam, October 8-12,1994 cal Hospital, Ministry of Health, Concepci6n, Chile; H. R. Wabinga, Makerere
(25) Koutsky LA, Holmes KK, Critchlow CW, et al: A cohort study of the risk Medical School, Kampala, Uganda; W. Zatonski, M. Sklodowska-Curie
of cervical intraepithelial neoplasia grade 2 or 3 in relation to papil- Memorial Cancer Centre, Warszawa, Poland; B. Sylla, P. Vizcaino, D. Magnin,
lomavirus infection. N Engl J Med 327:1272-1278, 1992 International Agency for Research on Cancer, Lyon, France; J. Kaldor, National
(26) Parkin DM, Pisani P, Ferlay J: Estimates of the worldwide incidence of Centre in HIV Epidemiology and Clinical Research, Darlinghurst, Australia; C.
802 ARTICLES Journal of the National Cancer Institute, Vol. 87, No. 11, June 7,1995