GYNECOLOGICONCOLOGY40, 25-28 (1991)
Invasive Carcinoma after Cone Biopsy for Cervical lntraepithelial Neoplasia JOHN Depurtment
of Obstetrics Consortium.
V. BROWN, M.D.,’ WILLIAM A. PETERS, M.D.,
und Gynecology, University of Wushington Seattle, Washington 98104; and Laboratory
AND DAVID
J.
CORWIN,
M.D.
School qf Medicine. Scuttle, Washington 98195; Puget Sound of Pathology of Seattle, Inc., Seattle, Washington 98104
Oncology
Received June 29, 1990
MATERIAL
A retrospective review of 578 patients with invasive cervical
cancer identified 8 patients with a history of one or more cone biopsies for treatment of cervical intraepithelial neoplasia. The cone biopsy and hysterectomy specimenswere reviewed to identify factors predictive of the subsequent development of invasive cancer. The mean interval from cone biopsy to diagnosis of invasive carcinoma was 6.7 years (range 1.5-16.5 years). High-grade intraepithelial neoplasia is a potentially invasive lesion. Adenocarcinema in situ was identified as a high-risk lesion that may be inadequately treated by cone biopsy. Four patients developed invasive squamous cancer in spite of complete excision of the initial lesion. Patients who have high-grade intraepithelial neoplasia treated with cone biopsy require long-term follow-up, and conization may hamper the subsequent diagnosis of preinvasive lesions. 0 1991AcademicPres.s,Inc.
AND METHODS
The tumor registries from two member institutions of the Puget Sound Oncology Consortium (Swedish Hospital Medical Center and Virginia Mason Medical Center) were reviewed. Five hundred and seventy-eight patients were treated for invasive carcinoma of the cervix at the two hospitals from January 1, 1976, to December 31, 1989. Nine patients were found to have one or more cone biopsies for the treatment of cervical intraepithelial neoplasia prior to the development of invasive cancer. One patient was excluded because invasive carcinoma was present on the initial cone biopsy. The histology was reviewed by a pathologist (D.C.). Each cone biopsy was evaluated for the grade of cervical intraepithelial neoplasia, the presence of invasion, and the involvement of margins. The specimens from the three extrafascial hysterectomies and five radical hysterectomies were evaluated for the same criteria.
INTRODUCTION Cone biopsy of the cervix is used to diagnose and treat cervical intraepithelial neoplasia. Despite regular cytologic follow-up after a cone biopsy performed for intraepithelial neoplasia, some patients develop invasive carcinoma of the cervix [l]. Kolstad and Klem reported that 0.9% of patients developed invasive carcinoma of the cervix after a cone biopsy for squamous carcinoma in situ [2]. Which patients are at highest risk to develop invasive carcinoma after cone biopsy is unclear. We found eight patients who had invasive cancer of the cervix after one or more previous cone biopsies for cervical intraepithelial neoplasia. A review of the pathology from the cone biopsies and hysterectomy specimens was undertaken to identify factors predictive of the subsequent development of invasive cervical carcinoma.
RESULTS
Age, histology, margin involvement, and interval from the first cone biopsy to the diagnosis of cancer for the eight patients are presented in Table 1. Seven patients had squamous cervical intraepithelial neoplasia III (highgrade cervical intraepithelial neoplasia) on the initial cone biopsy [3]. One of these also had adenocarcinoma in situ. One patient had squamous cervical intraepithelial neoplasia II (high-grade cervical intraepithelial neoplasia) on the initial cone biopsy [3]. Four margins were negative and three were positive. In one patient, margins could not be evaluated adequately. All patients had postconization Pap smears at regular intervals. No patient was lost to follow-up in the interval between cone biopsy and the development of invasive
’ To whom requests for reprints should be addressed at Suite 507, 351 Hospital Road, Newport Beach, CA 92663. 25
009%8258/9I $ I SO
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
26
BROWN, PETERS, AND CORWIN
TABLE 1 Treatment and Pathology Summary for Patients with a History of Cone Biopsy
Case Age
Diagnosis
Margin
Time from cone biopsy to diagnosis of cancer (years)
M.E.
46
CIN III”
Negative
16.5
R.G.
40
CIN III
B.H.
54
CIN II
Negative
1.6
C.H.
64
CIN III
Positive
3.6
M.S.
43
CIN III
Negative
11.7
K.S.
29
Pathology of first cone
h
Positive CIN III Adenocarcinoma
9.0
6.8
Pathology of second cone Diagnosis
Margin Treatment
Squamous cell carci- Positive noma Squamous cell carci- Positive noma Negative CIN I Squamous cell carci- Positive noma Squamous cell carci- Positive noma Positive Adenocarcinoma
RH RH TAH TAH RH RH
Pathology of hysterectomy Diagnosis
Margin
Stage
Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma No residual disease No residual disease Adenocarcinoma
Negative
IB
Negative
IB
Positive
’
Negative
IA,
Negative
IB
Negative
IA,
Negative
IB
Negative
IA,
in situ
in situ
D.W.
41
CIN III
Positive
1.5
CT.
45
CIN III
Negative
2.9
Squamous cell carci- Positive noma CIN III Positive
RH VH/RT
Squamous cell carcinoma Squamous cell carcinoma
” CIN, cervical intraepithelial neoplasia; RH, radical hysterectomy; TAH, total abdominal hysterectomy; VH, vaginal hysterectomy; RT, radiation therapy. ’ Unable to determine margin. ’ Carcinoma confined to cervix at hysterectomy.
FIG.
1.
Invasive cancer from B.H. buried in cervical stroma hysterectomy specimen.
CERVICAL
FIG. 2.
CANCER
AFTER
CONE
27
BIOPSY
Second cone biopsy from M.S. showing cancer eroding cervical
cancer. The mean interval from initial cone biopsy to the diagnosis of invasive cancer was 6.7 years (range 1S-16.5 years). Invasive carcinoma was diagnosed on the second cone biopsy in six of eight patients. One patient underwent an extrafascial hysterectomy for persistent abnormal cytology. Pathology revealed invasive squamous cell carcinoma within the cervical stroma. Another patient underwent a vaginal hysterectomy following a cone biopsy for cervical intraepithelial neoplasia III. Histology from the hysterectomy demonstrated an invasive cervical cancer with lymph-vascular channel involvement. Seven
epithelium.
patients had a squamous cell carcinoma and one had an adenocarcinoma. All eight patients had carcinoma confined to the cervix at the time of diagnosis. Four patients were FIG0 stage IB and three patients were stage IA2 [4]. One patient had carcinoma limited to the cervix at the time of laparotomy. DISCUSSION The development of cervical cancer after a cone biopsy for intraepithelial neoplasia is rare [ 1,231. McIndoe et
28
BROWN, PETERS, AND CORWIN
al. found that 1.5% of patients with normal Pap smears after treatment for cervical intraepithelial neoplasia III developed invasive cancer [ 11.In the present series, 1.4% of the patients with cervical cancer had a cone biopsy for intraepithelial lesions. Three patients had invasive cancer after apparent complete excision of the original lesion. The recurrence of high-grade cervical intraepithelial neoplasia and the subsequent development of invasive cancer after a cone biopsy for intraepithelial disease have been noted by other authors [I ,231. The mean interval of 6.7 years from cone biopsy to invasive cancer may reflect a new lesion or the natural history of intraepithelial neoplasia. A long duration of preinvasive disease was found by Kolstad and Klem, who recommended at least a IO-year follow-up after treatment for cervical intraepithelial neoplasia III [2]. One patient in their series developed invasive cancer 11 years after the original cone biopsy. Two patients had lesions buried in the cervical stroma which prevented preinvasive diagnosis. One patient (B.H.) had a lesion deep in the cervical stroma at hysterectomy that was inaccessible by both cone biopsies (Fig. 1). The other patient (M.S.) had a lesion invading from the stroma onto the cervical portio (Fig. 2). These lesions may represent abnormal epithelium that was buried by the original cone biopsy. Adenocarcinoma in situ of the cervix is rarely seen [6]. Due to the infrequency of the lesion, it is difficult to know the natural history. Our patient demonstrates the invasive potential of this lesion as shown by other investigators [6-81. Cone biopsy may not be adequate therapy for adenocarcinoma in situ because of difficulty evaluating margins, the multifocal nature of the disease, and disease in the deep endocervical glands. The authors recognize the potential limitations of a
retrospective review. In this study, the total number of cone biopsies performed could not be determined because only four patients underwent both cone biopsies at one of the institutions participating in this study. The other four patients were referred from multiple hospitals to a gynecologic oncologist at the time of or shortly following the second cone biopsy. As shown in previous studies, complete excision and normal follow-up Pap smears do not prevent invasive cervical carcinoma [1,2,5]. Adenocarcinoma in situ was also identified as a high-risk lesion that may be inadequately treated by cone biopsy. Our study suggests that long-term follow-up is necessary for patients with highgrade cervical intraepithelial neoplasia treated with cone biopsy. Anatomic distortion of the cervix after cone biopsy may hamper the detection of a preinvasive lesion. REFERENCES I. McIndoe, W. A., McLean, M. R., Jones, R. W., and Mullins, P. R. The invasive potential of carcinoma in situ of the cervix, Obstet. Gynecol. 64, 451-458 (1984). 2. Kolstad, P., and Klem, V. Long-term follow-up of 1121 cases of carcinoma in situ, Obstet. Gynecol. 48, 125-129 (1976). 3. Richart, R. M. A modified terminology for cervical intraepithelial neoplasia, Obstet. Gynecol. 75, 131-133 (1990). 4. Annu. Rep. Gynecol. Cancer FIG0 20, 30-33 (1988). 5. Creasman, W. T., and Rutledge, F. Carcinoma in situ of the cervix. An analysis of 861 patients, Obstet. Gynecol. 39, 373-380 (1972). 6. Christopherson, W. M., Nealon, N., and Gray, L. A. Noninvasive precursor lesions of adenocarcinoma and mixed adenosquamous carcinoma of the cervix uteri, Cancer 44, 975-983 (1979). 7. Abell, M. R., and Gosling, J. G. Gland cell carcinoma (adenocarcinema) of uterine cervix, Amer. J. Obstet. Gynacol. 83, 729-755 (1962). 8. Davis, J. R., and Moon, L. B. Increased incidence of adenocarcinoma of uterine cervix, Obstet. Gynecol. 45, 79-83 (1975).
Invasive Carcinoma after Cone Biopsy for Cervical lntraepithelial Neoplasia JOHN Depurtment
of Obstetrics Consortium.
V. BROWN, M.D.,’ WILLIAM A. PETERS, M.D.,
und Gynecology, University of Wushington Seattle, Washington 98104; and Laboratory
AND DAVID
J.
CORWIN,
M.D.
School qf Medicine. Scuttle, Washington 98195; Puget Sound of Pathology of Seattle, Inc., Seattle, Washington 98104
Oncology
Received June 29, 1990
MATERIAL
A retrospective review of 578 patients with invasive cervical
cancer identified 8 patients with a history of one or more cone biopsies for treatment of cervical intraepithelial neoplasia. The cone biopsy and hysterectomy specimenswere reviewed to identify factors predictive of the subsequent development of invasive cancer. The mean interval from cone biopsy to diagnosis of invasive carcinoma was 6.7 years (range 1.5-16.5 years). High-grade intraepithelial neoplasia is a potentially invasive lesion. Adenocarcinema in situ was identified as a high-risk lesion that may be inadequately treated by cone biopsy. Four patients developed invasive squamous cancer in spite of complete excision of the initial lesion. Patients who have high-grade intraepithelial neoplasia treated with cone biopsy require long-term follow-up, and conization may hamper the subsequent diagnosis of preinvasive lesions. 0 1991AcademicPres.s,Inc.
AND METHODS
The tumor registries from two member institutions of the Puget Sound Oncology Consortium (Swedish Hospital Medical Center and Virginia Mason Medical Center) were reviewed. Five hundred and seventy-eight patients were treated for invasive carcinoma of the cervix at the two hospitals from January 1, 1976, to December 31, 1989. Nine patients were found to have one or more cone biopsies for the treatment of cervical intraepithelial neoplasia prior to the development of invasive cancer. One patient was excluded because invasive carcinoma was present on the initial cone biopsy. The histology was reviewed by a pathologist (D.C.). Each cone biopsy was evaluated for the grade of cervical intraepithelial neoplasia, the presence of invasion, and the involvement of margins. The specimens from the three extrafascial hysterectomies and five radical hysterectomies were evaluated for the same criteria.
INTRODUCTION Cone biopsy of the cervix is used to diagnose and treat cervical intraepithelial neoplasia. Despite regular cytologic follow-up after a cone biopsy performed for intraepithelial neoplasia, some patients develop invasive carcinoma of the cervix [l]. Kolstad and Klem reported that 0.9% of patients developed invasive carcinoma of the cervix after a cone biopsy for squamous carcinoma in situ [2]. Which patients are at highest risk to develop invasive carcinoma after cone biopsy is unclear. We found eight patients who had invasive cancer of the cervix after one or more previous cone biopsies for cervical intraepithelial neoplasia. A review of the pathology from the cone biopsies and hysterectomy specimens was undertaken to identify factors predictive of the subsequent development of invasive cervical carcinoma.
RESULTS
Age, histology, margin involvement, and interval from the first cone biopsy to the diagnosis of cancer for the eight patients are presented in Table 1. Seven patients had squamous cervical intraepithelial neoplasia III (highgrade cervical intraepithelial neoplasia) on the initial cone biopsy [3]. One of these also had adenocarcinoma in situ. One patient had squamous cervical intraepithelial neoplasia II (high-grade cervical intraepithelial neoplasia) on the initial cone biopsy [3]. Four margins were negative and three were positive. In one patient, margins could not be evaluated adequately. All patients had postconization Pap smears at regular intervals. No patient was lost to follow-up in the interval between cone biopsy and the development of invasive
’ To whom requests for reprints should be addressed at Suite 507, 351 Hospital Road, Newport Beach, CA 92663. 25
009%8258/9I $ I SO
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
26
BROWN, PETERS, AND CORWIN
TABLE 1 Treatment and Pathology Summary for Patients with a History of Cone Biopsy
Case Age
Diagnosis
Margin
Time from cone biopsy to diagnosis of cancer (years)
M.E.
46
CIN III”
Negative
16.5
R.G.
40
CIN III
B.H.
54
CIN II
Negative
1.6
C.H.
64
CIN III
Positive
3.6
M.S.
43
CIN III
Negative
11.7
K.S.
29
Pathology of first cone
h
Positive CIN III Adenocarcinoma
9.0
6.8
Pathology of second cone Diagnosis
Margin Treatment
Squamous cell carci- Positive noma Squamous cell carci- Positive noma Negative CIN I Squamous cell carci- Positive noma Squamous cell carci- Positive noma Positive Adenocarcinoma
RH RH TAH TAH RH RH
Pathology of hysterectomy Diagnosis
Margin
Stage
Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma No residual disease No residual disease Adenocarcinoma
Negative
IB
Negative
IB
Positive
’
Negative
IA,
Negative
IB
Negative
IA,
Negative
IB
Negative
IA,
in situ
in situ
D.W.
41
CIN III
Positive
1.5
CT.
45
CIN III
Negative
2.9
Squamous cell carci- Positive noma CIN III Positive
RH VH/RT
Squamous cell carcinoma Squamous cell carcinoma
” CIN, cervical intraepithelial neoplasia; RH, radical hysterectomy; TAH, total abdominal hysterectomy; VH, vaginal hysterectomy; RT, radiation therapy. ’ Unable to determine margin. ’ Carcinoma confined to cervix at hysterectomy.
FIG.
1.
Invasive cancer from B.H. buried in cervical stroma hysterectomy specimen.
CERVICAL
FIG. 2.
CANCER
AFTER
CONE
27
BIOPSY
Second cone biopsy from M.S. showing cancer eroding cervical
cancer. The mean interval from initial cone biopsy to the diagnosis of invasive cancer was 6.7 years (range 1S-16.5 years). Invasive carcinoma was diagnosed on the second cone biopsy in six of eight patients. One patient underwent an extrafascial hysterectomy for persistent abnormal cytology. Pathology revealed invasive squamous cell carcinoma within the cervical stroma. Another patient underwent a vaginal hysterectomy following a cone biopsy for cervical intraepithelial neoplasia III. Histology from the hysterectomy demonstrated an invasive cervical cancer with lymph-vascular channel involvement. Seven
epithelium.
patients had a squamous cell carcinoma and one had an adenocarcinoma. All eight patients had carcinoma confined to the cervix at the time of diagnosis. Four patients were FIG0 stage IB and three patients were stage IA2 [4]. One patient had carcinoma limited to the cervix at the time of laparotomy. DISCUSSION The development of cervical cancer after a cone biopsy for intraepithelial neoplasia is rare [ 1,231. McIndoe et
28
BROWN, PETERS, AND CORWIN
al. found that 1.5% of patients with normal Pap smears after treatment for cervical intraepithelial neoplasia III developed invasive cancer [ 11.In the present series, 1.4% of the patients with cervical cancer had a cone biopsy for intraepithelial lesions. Three patients had invasive cancer after apparent complete excision of the original lesion. The recurrence of high-grade cervical intraepithelial neoplasia and the subsequent development of invasive cancer after a cone biopsy for intraepithelial disease have been noted by other authors [I ,231. The mean interval of 6.7 years from cone biopsy to invasive cancer may reflect a new lesion or the natural history of intraepithelial neoplasia. A long duration of preinvasive disease was found by Kolstad and Klem, who recommended at least a IO-year follow-up after treatment for cervical intraepithelial neoplasia III [2]. One patient in their series developed invasive cancer 11 years after the original cone biopsy. Two patients had lesions buried in the cervical stroma which prevented preinvasive diagnosis. One patient (B.H.) had a lesion deep in the cervical stroma at hysterectomy that was inaccessible by both cone biopsies (Fig. 1). The other patient (M.S.) had a lesion invading from the stroma onto the cervical portio (Fig. 2). These lesions may represent abnormal epithelium that was buried by the original cone biopsy. Adenocarcinoma in situ of the cervix is rarely seen [6]. Due to the infrequency of the lesion, it is difficult to know the natural history. Our patient demonstrates the invasive potential of this lesion as shown by other investigators [6-81. Cone biopsy may not be adequate therapy for adenocarcinoma in situ because of difficulty evaluating margins, the multifocal nature of the disease, and disease in the deep endocervical glands. The authors recognize the potential limitations of a
retrospective review. In this study, the total number of cone biopsies performed could not be determined because only four patients underwent both cone biopsies at one of the institutions participating in this study. The other four patients were referred from multiple hospitals to a gynecologic oncologist at the time of or shortly following the second cone biopsy. As shown in previous studies, complete excision and normal follow-up Pap smears do not prevent invasive cervical carcinoma [1,2,5]. Adenocarcinoma in situ was also identified as a high-risk lesion that may be inadequately treated by cone biopsy. Our study suggests that long-term follow-up is necessary for patients with highgrade cervical intraepithelial neoplasia treated with cone biopsy. Anatomic distortion of the cervix after cone biopsy may hamper the detection of a preinvasive lesion. REFERENCES I. McIndoe, W. A., McLean, M. R., Jones, R. W., and Mullins, P. R. The invasive potential of carcinoma in situ of the cervix, Obstet. Gynecol. 64, 451-458 (1984). 2. Kolstad, P., and Klem, V. Long-term follow-up of 1121 cases of carcinoma in situ, Obstet. Gynecol. 48, 125-129 (1976). 3. Richart, R. M. A modified terminology for cervical intraepithelial neoplasia, Obstet. Gynecol. 75, 131-133 (1990). 4. Annu. Rep. Gynecol. Cancer FIG0 20, 30-33 (1988). 5. Creasman, W. T., and Rutledge, F. Carcinoma in situ of the cervix. An analysis of 861 patients, Obstet. Gynecol. 39, 373-380 (1972). 6. Christopherson, W. M., Nealon, N., and Gray, L. A. Noninvasive precursor lesions of adenocarcinoma and mixed adenosquamous carcinoma of the cervix uteri, Cancer 44, 975-983 (1979). 7. Abell, M. R., and Gosling, J. G. Gland cell carcinoma (adenocarcinema) of uterine cervix, Amer. J. Obstet. Gynacol. 83, 729-755 (1962). 8. Davis, J. R., and Moon, L. B. Increased incidence of adenocarcinoma of uterine cervix, Obstet. Gynecol. 45, 79-83 (1975).
