Annals of Oncology 21: 994–1000, 2010

doi:10.1093/annonc/mdp426

Published online 25 October 2009original article

Management of occult invasive cervical cancer foundafter simple hysterectomy

J.-Y. Park, D.-Y. Kim, J.-H. Kim, Y.-M. Kim, Y.-T. Kim & J.-H. Nam*Department of Obstetrics and Gynecology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea

Received 28 May 2009; revised 14 July 2009; accepted 29 July 2009

Background: To estimate safety and efficacy of radical parametrectomy (RP) and radiation therapy (RT) or

concurrent chemoradiation therapy (CCRT) for patients with occult invasive cervical cancer found after simple

hysterectomy.

Materials and methods: We retrospectively evaluated outcomes in 147 patients with occult invasive cervical

cancer.

Results: Forty-eight patients with IA1 lesions (IA1 group) did not receive further treatment. Of the 99 patients with

IA2–IIA lesions, 26 received no definitive treatment (observation group), 44 received RT or CCRT (RT/CCRT group),

and 29 underwent RP (RP group). After a median follow-up of 116 months (range 3–235 months), recurrent disease

was observed in 0%, 34.6%, 6.8%, and 0% of patients in the IA1, observation, RT/CCRT, and RP groups,

respectively. In the RT/CCRT group, treatment was delayed due to severe diarrhea in 4 patients (9%) and 12 patients

(27%) had late complications related to RT requiring further management (including two surgical interventions). Five

patients in the RP group (17%) experienced perioperative complications which were easily managed, intraoperatively

or conservatively. Late complications were not observed in the RP group.

Conclusion: Although RP and RT/CCRT had similar therapeutic efficacy, the lower rate of late complications

observed with RP makes it preferable to RT/CCRT.

Key words: concurrent chemoradiation therapy, occult cervical cancer, radiation therapy, radical parametrectomy,

simple hysterectomy

introduction

Cervical cancer is the second most common female cancer andone of the leading causes of cancer deaths in females worldwide[1, 2]. Widespread screening for cervical cancer has increasedthe rate of early-stage diagnosis. Most patients with early-stagecervical cancer undergo radical hysterectomy with pelvic 6

paraaortic lymphadenectomy, with 5-year survival rates of75%–90% [3–5]. Sometimes, however, this malignancy isencountered after simple hysterectomy carried out for benigngynecologic conditions or preinvasive cervical lesions. Radicalparametrectomy (RP), consisting of resection of theparametrium, upper vaginectomy, and pelvic 6 paraaorticlymphadenectomy may be carried out as a definite treatment inthese patients [6, 7]. However, due to technical difficulties incarrying out RP and a lack of knowledge about the safety andefficacy of this operation, physicians tend to administerradiation therapy (RT) or concurrent chemoradiation therapy(CCRT) instead.

The aims of this study were to estimate the feasibility, safety,and efficacy of RP for occult invasive cervical cancer detectedafter simple hysterectomy carried out for benign gynecologicconditions or preinvasive cervical lesions and to compare theoutcomes of patients who underwent RP with those whoreceived RT or CCRT.

materials and methods

Following approval by the Institutional Review Board of the Asan Medical

Center (AMC, Seoul, Korea), we searched the cancer registry and

computerized database to identify patients with occult invasive cervical

cancer detected after simple hysterectomy carried out for benign

gynecologic conditions or preinvasive cervical lesions from 1989 to 2009.

Patients’ medical records were retrospectively reviewed, and demographic

data including age, menopausal status, parity, and body mass index were

recorded. We also reviewed patient’s medical records for a history of cancer,

other medical diseases, and surgery or RT of the pelvis. Clinical data

included indication for hysterectomy; results of Papanicolaou smears taken

before hysterectomy; presence of residual tumor; histologic type, size, and

grade of each tumor; presence of lymphovascular space invasion (LVSI) in

the hysterectomy specimen; presumed stage of disease, as determined using

the International Federation of Gynecology and Obstetrics (FIGO) system

for cervical cancer; the outcomes of diagnostic procedures; type and

ori

gin

al

art

icle

*Correspondence to: Dr J.-H. Nam, Department of Obstetrics and Gynecology,

University of Ulsan College of Medicine, Asan Medical Center, 388-1 Poongnap-2 Dong,

Songpa-Gu, Seoul 138-736, Korea. Tel: +82-2-3010-3633; Fax: +82-2-476-7331;

E-mail: jhnam@amc.seoul.kr

ª The Author 2009. Published by Oxford University Press on behalf of the European Society for Medical Oncology.

All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org

by guest on October 4, 2015

http://annonc.oxfordjournals.org/D

ownloaded from

outcomes of definitive treatment after hysterectomy; lymph node (LN)

status after RP; adjuvant therapy after RP; recurrence; and treatment at

recurrence and death. Pathology slides were reviewed by two experienced

pathologists and clinicopathological prognostic factors and treatment

outcomes were analyzed.

surgical procedure for RPRP surgery was carried out through a midline laparotomy in 25 patients

and through laparoscopic surgery using four trocars in 4 patients. Both

surgical approaches were basically the same with type III hysterectomy [8].

Briefly, each surgical procedure commenced with a systemic bilateral pelvic

6 paraaortic lymphadenectomy. The ureters were dissected from the

uterine artery and the cardinal ligaments and ureters were separated as

much as possible from the bladder. Cardinal ligaments were severed at their

most lateral portions and freed as much as possible from the surrounding

tissue. The uterosacral ligaments were severed near the sacrum, and the

paracolpium was resected as much as possible. An incision was made in the

vaginal cuff at least 3–4 cm below the fornices.

statistical analysisFrequency distributions were compared using the chi-square and Fisher’s

exact tests, and mean and median values were compared between the

groups using the Student’s t-test and the Mann–Whitney U test. Overall

survival (OS) was calculated as the number of months from the date of

simple hysterectomy to the date of death or the date censored. Disease-free

survival (DFS) was calculated as the number of months from the date of

simple hysterectomy to the date of recurrence or the date censored. Survival

curves and rates were calculated using the Kaplan–Meier method.

Differences in survival were assessed using the log-rank test for categorical

factors and Cox proportional hazards model for continuous factors in

univariate analysis. A P value of <0.05 in a two-sided test indicated

a significant difference. Data were analyzed using SPSS software for

Windows (version 11.0; SPSS Inc., Chicago, IL).

results

Of 2792 patients with invasive cervical cancer who were treatedand followed at AMC during the study period, 147 (5.3%) hadoccult invasive cervical cancer found after simple hysterectomycarried out for benign gynecologic conditions or preinvasivecervical lesions. Of these 147 patients, 37 (25.2%) hadundergone simple hysterectomy at other hospitals and 110(74.8%) had undergone simple hysterectomy at AMC. Theindications for simple hysterectomy included squamousintraepithelial neoplasia (CIN), including carcinoma in situ(CIS), in 102 patients (69.4%), leiomyoma in 32 patients(21.8%), adenomyosis in 4 patients (2.7%), adenocarcinoma insitu in 4 patients (2.7%), uterine prolapse in 3 patients (2.0%),intractable dysfunctional uterine bleeding in 1 patient (0.7%),and severe postpartum hemorrhage in 1 patient (0.7%). Of102 patients who underwent inadvertent simple hysterectomyfor CIN or CIS, 31 did so due to the failure to carry outconization or endocervical curettage and 71 did so due to theabsence of invasive lesions in conization specimens. Of theremaining 45 patients who underwent inadvertent simplehysterectomy due to other benign gynecologic problems, 35 didso due to false-negative cytology and 10 did so due to theabsence of cytologic evaluation before hysterectomy. Ninety-five patients (64.6%) underwent total abdominal hysterectomy,47 (32.0%) underwent laparoscopic-assisted vaginal

hysterectomy, and 5 (3.4%) underwent vaginal hysterectomy.Sixty-five patients underwent unilateral (n = 11) or bilateral(n = 51) salpingo-oophorectomy or ovarian cystectomy(n = 3) during hysterectomy. After simple hysterectomy,presumed FIGO stage was IA1 in 48 patients (32.7%), IA2 in7 patients (4.8%), IB1 in 85 patients (57.8%), IB2 in 4 patients(2.7%), and IIA in 3 patients (2.0%). The characteristics of the147 patients are shown in Table 1.

patients with IA1 lesions

The hysterectomy specimens from all 48 patients with IA1lesions had negative resection margins. All patients hadsquamous lesions. After simple hysterectomy, none underwentfurther imaging. Six patients (12.5%) had positive LVSI. Noneof these patients underwent further treatment after simplehysterectomy. After a median follow-up time of 158 months(range 34–235 months), none of these patients had recurrentdisease (Figure 1).

patients with IA2–IIA lesions

Of the 99 patients with IA2–IIA lesions, 26 received no furtherdefinitive treatment, including RT, CCRT, or RP (observation/chemotherapy group), and 44 patients received RT or CCRT(RT/CCRT group) and 29 underwent RP (RP group) asdefinitive treatments after simple hysterectomy.

observation/chemotherapy group. Of the 26 patients who did notreceive RT, CCRT, or RP, 20 refused further treatment, whereas6 received adjuvant chemotherapy; two patients receivedpaclitaxel/cisplatin, three received 5-fluorouracil/cisplatin, andone received vincristine/ifosfamide/cisplatin. The meannumber of chemotherapy cycles was 4 (range 3–6). All sixpatients who received adjuvant chemotherapy had IB1 lesions;of these, three had squamous cell carcinoma and three hadadenocarcinoma. Of the 26 patients in this group, 2 underwentcomputed tomography (CT) of the abdomen and pelvis, 3underwent magnetic resonance imaging (MRI) of the abdomenand pelvis, and 2 underwent whole-body positron emissiontomography–computed tomography (PET–CT) after simplehysterectomy. None showed evidence of a pelvic lesion orlymphadenopathy. After a median follow-up time of 104months (range 7–232 months), nine patients (34.6%) hadrecurrent disease. The median time to recurrence was 46months (range 4–137 months). Recurrent sites were the vaginalstump (n = 2), the pelvis (n = 4), the pelvis and pelvic LNs(n = 1), and the pelvis and paraaortic LNs (n = 2). Fourpatients received RT and five patients received CCRT atrecurrence, with four dying of disease. The 10-year DFS and OSrates were 63% and 84%, respectively (Figure 1). Of the sixpatients who received adjuvant chemotherapy, one (17%) hadrecurrent disease and died of disease. Of the remaining 20patients who did not receive adjuvant chemotherapy, 8 (40%)had recurrent disease and 3 (15%) died of disease. There wereno differences in DFS and OS between the two groups(P = 0.315 and 0.935, respectively).

RT/CCRT group. Thirty-two patients received RT and 12received CCRT. The chemotherapeutic regimen was weekly

Annals of Oncology original article

Volume 21 | No. 5 | May 2010 doi:10.1093/annonc/mdp426 | 995

by guest on October 4, 2015

http://annonc.oxfordjournals.org/D

ownloaded from

cisplatin in one patient, paclitaxel/cisplatin in two patients, and5-fluorouracil/cisplatin in nine patients. The mean number ofchemotherapy cycles was 3.3 (range 1–6). The mean time fromhysterectomy to the commencement of RT or CCRT was 29days (range 8–124 days). Twenty-three patients received whole-pelvic radiation therapy (WPRT) only, 3 patients receivedintracavitary radiation therapy (ICR) only, and 18 patientsreceived both. The mean radiation dose for WPRT was 5006cGy (range 4000–5080 cGy), and the mean radiation dose forICR was 2162 cGy (range 1500–3000 cGy). The mean durationof RT or CCRT was 49 days (range 13–67 days). In fourpatients, RT or CCRT was delayed due to severe diarrhea. RT

or CCRT were well tolerated in other patients. Of the 44patients in this group, 7 underwent CT, 12 underwent MRI,and 2 underwent PET–CT after simple hysterectomy. Onlytwo MRI scans revealed suspected residual tumors in thevaginal stump. After a median follow-up time of 116 months(range 9–232 months), three patients (6.8%) had recurrentdisease. The median time to recurrence was 9 months (range6–19 months). Recurrent sites were the pelvis in one patient;the pelvis, paraaortic, and cervical LNs in one patient; andbone and lung in one patient. One patient receivedchemotherapy consisting of three cycles of paclitaxel/cisplatin,and two patients received CCRT (one received three cycles of

Table 1. Characteristics of patients (N = 147)

Characteristics Total IA2–IIA lesiona P valueb

Obs or CTx RT or CCRT RP

Number of patients, n (%) 147 (100.0) 26 (17.7) 44 (29.9) 29 (19.7)

Age (years), mean (range) 48 (28–75) 45 (28–63) 52 (34–75) 51 (33–74) 0.776b

Menopause, n (%) 0.535b

No 78 (53.1) 15 (57.7) 18 (40.9) 14 (48.3)

Yes 69 (46.9) 11 (42.3) 26 (59.1) 15 (51.7)

Parity, n (%) 0.008b

£2 78 (53.1) 15 (57.7) 15 (34.1) 19 (65.5)

>2 69 (46.9) 11 (42.3) 29 (65.9) 10 (34.5)

FIGO stage, n (%) 0.170b

IA1 48 (32.7) 0 (0.0) 0 (0.0) 0 (0.0)

IA2 7 (4.8) 5 (19.2) 5 (4.5) 0 (0.0)

IB1 85 (57.8) 21 (80.8) 37 (84.1) 27 (93.1)

IB2 4 (2.7) 0 (0.0) 9 (9.1) 0 (0.0)

IIA 3 (2.0) 0 (0.0) 1 (2.3) 2 (6.9)

Histology, n (%) 0.003b

Squamous cell carcinoma 121 (82.3) 21 (80.8) 36 (81.8) 16 (55.2)

Adenocarcinoma 23 (15.6) 5 (19.2) 5 (11.4) 13 (44.8)

Adenosquamous cell carcinoma 3 (2.0) 0 (0.0) 3 (6.8) 0 (0.0)

Grade, n (%) 0.056b

Well differentiated 28 (19.0) 7 (26.9) 6 (13.6) 11 (37.9)

Moderately differentiated 110 (74.8) 16 (61.5) 34 (77.3) 16 (55.2)

Poorly differentiated 9 (6.1) 3 (11.5) 4 (9.1) 2 (6.9)

Size of tumor (cm), mean (range) 1.1 (0.1–5.0) 1.1 (0.5–2.5) 1.7 (0.5–5.0) 1.7 (0.6–3.5) 0.969b

Depth of invasion, n (%) 0.026b

<1/3 98 (66.7) 21 (80.8) 23 (52.3) 6 (20.7)

1/3 to 2/3 25 (17.0) 2 (7.7) 11 (25.0) 12 (41.4)

‡2/3 24 (16.3) 3 (11.5) 10 (22.7) 11 (37.9)

Resection margin, n (%) 0.642b

Negative 137 (93.2) 26 (100.0) 40 (90.9) 28 (96.6)

Positive 10 (6.8) 0 (0.0) 4 (9.1) 1 (3.4)

LVSI, n (%) 0.108b

Negative 128 (87.1) 25 (96.2) 34 (77.3) 27 (93.1)

Positive 19 (12.9) 1 (3.8) 10 (22.7) 2 (6.9)

Follow-up time (months), median (range) 116 (3–235) 104 (7–232) 100 (9–232) 73 (3–220) 0.068b

Recurrence, n (%) 12 (8.2) 9 (34.6) 3 (6.8) 0 (0.0) 0.272b

Death, n (%) 6 (4.1) 4 (15.4) 2 (4.5) 0 (0.0) 0.514b

10-year DFS (%) 91 63 93 100 0.199b

10-year OS (%) 96 84 94 100 0.276b

aThe details of patients with IA1 lesion are not shown here.bComparison between the RT/CCRT and RP groups.

Obs, observation; CTx, chemotherapy; RT, radiation therapy; CCRT, concurrent chemoradiation therapy; RP, radical parametrectomy; FIGO, International

Federation of Gynecology and Obstetrics; LVSI, lymphovascular space invasion; DFS, disease-free survival; OS, overall survival.

original article Annals of Oncology

996 | Park et al. Volume 21 | No. 5 | May 2010

by guest on October 4, 2015

http://annonc.oxfordjournals.org/D

ownloaded from

paclitaxel/cisplatin and the other received three cycles of5-fluorouracil/cisplatin); of these three patients, two died ofdisease. The characteristics of the three patients who hadrecurrent disease after RT or CCRT are shown in Table 2. The10-year DFS and OS rates were 93% and 94%, respectively(Figure 1). During follow-up after RT or CCRT, 12 patients(27%) suffered late complications related to RT requiringfurther management; these complications included radiationcystitis (n = 4), radiation proctitis or colitis (n = 6), radiationvaginitis (n = 4), rectovaginal fistula (n = 1), and ureteralstricture (n = 1). Two patients required surgical management.Of the 12 patients with late complications, 8 (67%) hadreceived WPRT and ICR, with a total dose of >6540 cGy, and 4(33%) had received WPRT only, with a total dose of 5040 cGy.Of the 44 patients who received RT or CCRT, four had positiveresection margins on simple hysterectomy specimens. Of these,three patients with FIGO stage IB1 lesions received RT, onepatient with a FIGO stage IB2 lesion received CCRT with threecycles of 5-fluorouracil and cisplatin, and one patient witha FIGO stage 1B1 lesion underwent RP. None showed evidenceof recurrent disease after treatment.

RP group. Twenty-nine patients underwent RP with pelviclymphadenectomy. Nineteen patients underwent paraaorticlymphadenectomy, 11 underwent unilateral or bilateralsalpingo-oophorectomy, and 3 underwent ovariantransposition during RP. Four patients underwent laparoscopicRP. Of the 29 patients who underwent RP, one patient witha IB1 lesion had positive resection margins on simplehysterectomy specimen. One patient underwent CT, 10underwent MRI, and 7 underwent PET–CT after simplehysterectomy. Only one MRI scan revealed a residual tumor inthe vaginal stump. The mean time from simple hysterectomy toRP was 34 days (range 13–114 days). The mean operating timewas 297 min (range 172–433 min), and the mean estimatedblood loss during surgery was 538 ml (range 200–1000 ml).Seventeen patients (59%) required perioperative transfusions,with a mean transfusion volume of 2.4 pints (range 1–5 pints).The mean preoperative and postoperative hemoglobin (Hb)

concentrations and perioperative Hb change were 13.1 gm/dl(range 11.9–14.3), 11.0 gm/dl (range 9.1–13.6), and 2.1 gm/dl(range 0.5–5.6), respectively. Three intraoperativecomplications were recorded, including one bladderperforation and two rectal serosa lacerations, all of which weremanaged with intraoperative suturing. After surgery, twopatients experienced complications, including one withmechanical ileus that was managed conservatively and one withurethrocutaneous fistula that was managed surgically. Themean time to return of bowel movements was 2.8 days (range1.0–4.0 days). The mean postoperative hospital stay was 17 days(range 7–43 days). In three patients, residual disease was foundon the vaginal stump after RP, measuring 0.4, 0.5, and 0.5 cm,respectively. No RP specimen showed involvement of theparametrium or resection margin. The mean number of total,paraaortic, and pelvic LNs retrieved were 36.8 (range 13–57),4.3 (range 3–7), and 33.4 (13–53), respectively. Four patientshad LN metastasis; the mean number of LNs involved was 1.8(range 1–3). Two patients had right obturator LN metastasis,one had right internal iliac LN metastasis, and one hadinvolvement of both external iliac LNs. The characteristics ofthe four patients with LN metastasis are shown in Table 3. Fourpatients with LN metastasis and one with two intermediate riskfactors (positive LVSI and stromal invasion >2/3) receivedadjuvant therapy after RP; one received RT, one received fivecycles of paclitaxel/cisplatin, one received three cycles of 5-fluorouracil/cisplatin, and two received CCRT, consisting ofthree cycles of 5-fluorouracil/cisplatin or six cycles of weeklycisplatin. After a median follow-up of 73 months (range 3–220months), no patient showed evidence of disease recurrence(Figure 1) or late complications related to RP that requiredfurther management.

discussion

Our findings indicate that occult IA1 cervical cancer found aftersimple hysterectomy can be followed safely without furthermanagement, regardless of the status of LVSI. In patients with

Figure 1. Disease-free survival (left) and overall survival (right) by stage and treatment modality in 147 patients with occult invasive cervical cancer. IA1

observation (Obs), 48 patients with IA1 lesions who did not receive further management; Obs/chemotherapy (CTx), 26 patients with IA2–IIA lesions who

did not receive further management or who received adjuvant chemotherapy; radiation therapy (RT)/concurrent chemoradiation therapy (CCRT), 44

patients with IA2–IIA lesions who received RT or CCRT; radical parametrectomy (RP), 29 patients with IA2–IIA lesions who underwent RP.

Annals of Oncology original article

Volume 21 | No. 5 | May 2010 doi:10.1093/annonc/mdp426 | 997

by guest on October 4, 2015

http://annonc.oxfordjournals.org/D

ownloaded from

Table 2. The characteristics of patients who had recurrence after RT or CCRT

Patient Age

(years)

FIGO

stage

Histology Grade Tumor

size (cm)

LVSI DSI RM Treatment RT type RT dose

(cGy)

RFS

(months)

OS

(months)

Status

1 46 IB1 SCCa 2 1.1 Positive 1/3 to 2/3 Negative CCRT wP 6 cycles WPRT + ICR 6540 19 56 DOD

2 45 IB2 SCCa 3 4.0 Positive >2/3 Negative CCRT FP 6 cycles WPRT + ICR 7440 6 9 DOD

3 47 IB1 SCCa 2 1.5 Negative £1/3 Negative RT WPRT 5040 9 36 AWD

RT, radiation therapy; CCRT, concurrent chemoradiation therapy; FIGO, International Federation of Gynecology and Obstetrics; LVSI, lymphovascular space invasion; DSI, depth of cervical stromal invasion;

RM, resection margin on simple hysterectomy specimen; RFS, recurrence-free survival; OS, overall survival; SCCa, squamous cell carcinoma; wP, weekly cisplatin; WPRT, whole-pelvic radiation therapy; ICR,

intracavitary radiation therapy; DOD, die of disease; FP, 5-fluorouracil + cisplatin; AWD, alive with disease.

Table 3. The characteristics of patients with pelvic LN metastasis after RP

Patient Age (years) FIGO

stage

Histology Grade Tumor size

(cm)

LVSI DSI RM Site of LN

metastasis

No. of LN

metastases

Adjuvant treatment

after RP

Recurrence OS (months) Status

1 45 IB1 AdCa 2 2.5 Negative >2/3 Negative Both external iliac 3 CCRT FP 3 cycles None 123 NED

2 66 IIA SCCa 1 2.7 Negative >2/3 Negative Right internal iliac 1 CCRT wP 6 cycles None 3 NED

3 48 IB1 AdCa 2 3.5 Positive >2/3 Negative Right obturator 2 TP 5 cycles None 63 NED

4 66 IIA SCCa 1 2.0 Negative >2/3 Negative Right obturator 1 RT None 93 NED

LN, lymph node; RP, radical parametrectomy; FIGO, International Federation of Gynecology and Obstetrics; LVSI, lymphovascular space invasion; DSI, depth of cervical stromal invasion; RM, resection margin

on simple hysterectomy specimen; OS, overall survival; AdCa, adenocarcinoma; CCRT, concurrent chemoradiation therapy; FP, 5-fluorouracil + cisplatin; NED, no evidence of disease; SCCa, squamous cell

carcinoma; wP, weekly cisplatin; TP, paclitaxel + cisplatin; RT, radiation therapy.

orig

inalartic

leA

nnals

of

Oncolo

gy

998

|P

ark

et

al.

Volu

me

21

|No.

5|M

ay

2010

by guest on October 4, 2015 http://annonc.oxfordjournals.org/ Downloaded from

more advanced early-stage lesions (stage IA2–IIA), however,a definitive treatment, such as RT, CCRT, or RP, is necessarybecause these patients are at increased risk of recurrence anddeath, although they received adjuvant chemotherapy aftersimple hysterectomy. Although similar survival outcomes wereobtained in the RT/CCRT and RP groups, the lower rate of latecomplications after RP makes the latter preferable comparedwith RT/CCRT in this patient population. RP was feasible in allpatients in our series and the operative parameters andcomplications were acceptable.

Cervical cancer may be found incidentally after simplehysterectomy carried out for benign gynecologic conditions orpreinvasive cervical lesions. The incidence of occult invasivecervical cancer, however, is not clear, although in the presentstudy it comprised 5.3% of all cervical cancers. Occult invasivecervical cancer after simple hysterectomy can occur due to severalcauses. We found that the absence of invasive lesions in theconization specimens from patients with CIN or CIS and false-negative cytology were the most common causes of inadvertentsimple hysterectomy. Other studies have found that the mostcommon causes of inadvertent simple hysterectomy were the lackof preoperative Pap smear, negative cytology, and no clinicalevidence of cancer, followed by inadequate evaluation of anabnormal Pap smear or cervical biopsy and failure to carry outconization or endocervical curettage [7, 9, 10].

If a lesion is found to be IA1 cervical cancer, furthermanagement is not required, regardless of the status of LVSI[11, 12]. However, if the lesion is larger, definitive RT, CCRT,or RP is required because higher recurrence and death rateshave been observed in patients who did not receive furthermanagement or who received adjuvant chemotherapy only.Although we found that patients who did not receive definitivetreatment had smaller tumors and a greater incidence ofsuperficial (<1/3) stromal invasion than those in the RT/CCRTand RP groups, 34.6% of patients had recurrent disease. All ofthese patients had pelvic failure and some also had distantfailure. Using RT or CCRT for recurrent disease, five patientswere successfully re-treated, but four patients died of disease.These results are in agreement with previous findings indicatingthat simple hysterectomy is not sufficient for patients withmore than microinvasive cervical carcinoma (stage IA1 disease)due to high recurrence and mediocre survival rates [13–15].

Five-year survival rates after RT for occult cervical cancerhave been found to range from 54% to 93% [9, 16–27](Table 4). In our series, 10-year DFS and OS rates after RT orCCRT were 93% and 94%, respectively. Factors associated withpoor outcome after RT or CCRT may include residual diseaseafter simple hysterectomy [17, 18, 21, 24], positive resectionmargins in hysterectomy specimens [18], deeper cervicalstromal invasion [3, 20], long time interval from hysterectomyto RT (>6 months) due to delayed RT [18, 27], and histologictype of adenocarcinoma [3].

In general, late complications of RT for cervical cancer havebeen reported in 5%–15% of patients [28]. Most early studies,however, included relatively short follow-up periods. In morerecent studies, with longer follow-up, significant latecomplication rates of 10%–16% [29, 30] and 16%–21% havebeen reported [31]. In our series, late complications wereobserved in 27% of patients after a median follow-up of 100

months. Surgical intervention was required in 4.5% of patients,similar to previously reported rates of 6%–7% [3, 21, 24]. Inagreement with previous results, most of our patients with latecomplications received WPRT and ICR, with total dose >6540cGy [3, 21, 24]. Although the role of brachytherapy incombination with WPRT is not yet clear, most patients treatedwith WPRT alone in our series did well. Considering the higherrate of complications, ICR may be safely omitted in selectedpatients.

Five-year survival rates after RP have been reported to rangefrom 67% to 96% [6, 7, 32–35] (Table 4). We observed 10-yearDFS and OS rates after RP of 100%, providing further evidenceof the safety of RP in patients with: (i) FIGO stage IA2–IIAtumors; (ii) squamous cell carcinoma, adenocarcinoma, oradenosquamous cell carcinoma; and (iii) tumors <4 cm indiameter, regardless of grade of tumor, depth of invasion, orLVSI. Although previous studies have indicated that all patientswith positive surgical margins on simple hysterectomyspecimens are poor candidates for RP [6], our results indicatethat if there is no parametrial involvement, RP can be carriedout safely in patients with small residual disease on the vaginalstump after simple hysterectomy. However, if there is a highprobability that a patient will receive adjuvant RT or CCRTafter RP, then RT or CCRT should be carried out instead of RPbecause the combination of radical surgery and RT is associatedwith a particularly high morbidity rate with no further survivalbenefit [5]. Indeed, large prospective trials are required to fullyestablish the patient eligibility criteria for RP.

Table 4. The outcomes by treatment modality in patients with occult

cervical cancer found after simple hysterectomy

Author Year N Treatment

modality

5-year OS

rate (%)

Cosbie [16] 1963 86 RT 54

Barber et al. [15] 1968 115 RP 32

Green et al. [32] 1969 30 RT 30

21 RP 61

Andras et al. [17] 1973 118 RT 89

Davy et al. [18] 1977 72 RT 77

Papavasiliou et al. [19] 1980 36 RT 89

Heller et al. [20] 1986 35 RT 67

Orr et al. [33] 1986 23 RP NR

Kinney et al. [34] 1992 27 RP 82

Chapman et al. [35] 1992 18 RP 89

Roman et al. [21] 1993 122 RT 65

Fang et al. [22] 1993 73 RT 67

Choi et al. [23] 1997 64 RT 76

Crane et al. [24] 1999 18 RT 93

Huerta Bahena et al. [25] 2003 59 RT 59

Chen et al. [26] 2003 29 RT 82–95

Munstedt et al. [9] 2004 80 RT 83

Leath et al. [6] 2004 23 RP 96

Ayhan et al. [7] 2006 27 RP 89

Present study 2009 44 RT or CCRT 94a

29 RP 100a

a10-year OS rate.

OS, overall survival; RT, radiation therapy; RP, radical parametrectomy;

NR, not reported; CCRT, concurrent chemoradiation therapy.

Annals of Oncology original article

Volume 21 | No. 5 | May 2010 doi:10.1093/annonc/mdp426 | 999

by guest on October 4, 2015

http://annonc.oxfordjournals.org/D

ownloaded from

Most physicians are reluctant to carry out RP in patients withIA2–IIA occult invasive cervical cancer because they consider theRP procedure to be quite difficult, require an experiencedsurgeon, and be associated with a high rate of surgery-relatedmorbidity. Therefore, most patients with IA2–IIA occultinvasive cervical cancer undergo RT or CCRT [9, 16–27], withfew undergoing RP [6, 7, 32–35]. This is in contrast to thecurrent treatment of patients with early-stage cervical cancer,most of whom undergo radical hysterectomy as primarytreatment. RP has been found to be feasible for most patients andnot particularly technically difficult, as assessed by objectivemeasures such as operating time, estimated blood loss,transfusion requirement, and postoperative hospital stay [6, 7,32–35]. We found that the perioperative complication rate was17%, which is within the previously reported range of 8.7%–30%[6, 7, 32–35]. In the present study, most complications wereeasily managed, either intraoperatively or conservatively. Theoccurrence of long-term morbidity after RP is rare.

In conclusion, we have shown that the survival outcome afterRP was similar to that after RT or CCRT in patient populationswith similar disease stage, tumor size, positive resection margin,positive LVSI, grade of tumor, and depth of stromal invasion.In addition, RP was feasible in all patients. The immediatesurgical parameters were acceptable, the rate of perioperativecomplications was very low, and there was no late morbidity.Due to the high rates of long-term morbidity after RT orCCRT, RP may be preferable for selected patients with IA2–IIAoccult invasive cervical cancer. We think that RP may be ofgreatest benefit in young patients who want to preserve theirovarian and sexual function.

disclosure

There is no conflict of interest to declare.

references

1. Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 major

cancers in 1990. Int J Cancer 1999; 80: 827–841.

2. Pisani P, Parkin DM, Bray F, Ferlay J. Estimates of the worldwide mortality from

25 cancers in 1990. Int J Cancer 1999; 83: 18–29.

3. Hopkins MP, Morley GW. Radical hysterectomy versus radiation therapy for stage

IB squamous cell cancer of the cervix. Cancer 1991; 68: 272–277.

4. Burghardt E, Baltzer J, Tulusan AH, Haas J. Results of surgical treatment of

1028 cervical cancers studied with volumetry. Cancer 1992; 70: 648–655.

5. Landoni F, Maneo A, Colombo A et al. Randomised study of radical surgery versus

radiotherapy for stage Ib-IIa cervical cancer. Lancet 1997; 350: 535–540.

6. Leath CA III, Straughn JM, Bhoola SM et al. The role of radical parametrectomy

in the treatment of occult cervical carcinoma after extrafascial hysterectomy.

Gynecol Oncol 2004; 92: 215–219.

7. Ayhan A, Otegen U, Guven S, Kucukali T. Radical reoperation for invasive cervical

cancer found in simple hysterectomy. J Surg Oncol 2006; 94: 28–34.

8. Piver MS, Rutledge F, Smith JP. Five classes of extended hysterectomy for

women with cervical cancer. Obstet Gynecol 1974; 44: 265–272.

9. Munstedt K, Johnson P, von Georgi R et al. Consequences of inadvertent,

suboptimal primary surgery in carcinoma of the uterine cervix. Gynecol Oncol

2004; 94: 515–520.

10. Behtash N, Mousavi A, Mohit M et al. Simple hysterectomy in the presence of

invasive cervical cancer in Iran. Int J Gynecol Cancer 2003; 13: 177–181.

11. Morris M. Management of stage IA cervical carcinoma. J Natl Cancer Inst

Monogr 1996; 21: 47–52.

12. Schorge JO, Lee KR, Flynn CE et al. Stage IA1 cervical adenocarcinoma:

definition and treatment. Obstet Gynecol 1999; 93: 219–222.

13. Jones H, Jones G. Panhysterectomy versus irradiation in early cancer of the

cervix. JAMA 1943; 122: 930–932.

14. Daniel WW, Brunschwig A. The management of recurrent carcinoma of the

cervix following simple total hysterectomy. Cancer 1961; 14: 582–586.

15. Barber HR, Pece GV, Brunschwig A. Operative management of patients

previously operated upon for a benign lesion with cervical cancer as a surprise

finding. Am J Obstet Gynecol 1968; 101: 959–965.

16. Cosbie WG. Radiotherapy following hysterectomy performed for or in the

presence of cancer of the cervix. Am J Obstet Gynecol 1963; 85: 332–337.

17. Andras EJ, Fletcher GH, Rutledge F. Radiotherapy of carcinoma of the cervix

following simple hysterectomy. Am J Obstet Gynecol 1973; 115: 647–655.

18. Davy M, Bentzen H, Jahren R. Simple hysterectomy in the presence of invasive

cervical cancer. Acta Obstet Gynecol Scand 1977; 56: 105–108.

19. Papavasiliou C, Yiogarakis D, Pappas J, Keramopoulos A. Treatment of cervical

carcinoma by total hysterectomy and postoperative external irradiation. Int J

Radiat Oncol Biol Phys 1980; 6: 871–874.

20. Heller PB, Barnhill DR, Mayer AR et al. Cervical carcinoma found incidentally in

a uterus removed for benign indications. Obstet Gynecol 1986; 67: 187–190.

21. Roman LD, Morris M, Mitchell MF et al. Prognostic factors for patients

undergoing simple hysterectomy in the presence of invasive cancer of the cervix.

Gynecol Oncol 1993; 50: 179–184.

22. Fang FM, Yeh CY, Lai YL et al. Radiotherapy following simple hysterectomy in

patients with invasive carcinoma of the uterine cervix. J Formos Med Assoc

1993; 92: 420–425.

23. Choi DH, Huh SJ, Nam KH. Radiation therapy results for patients undergoing

inappropriate surgery in the presence of invasive cervical carcinoma. Gynecol

Oncol 1997; 65: 506–511.

24. Crane CH, Schneider BF. Occult carcinoma discovered after simple hysterectomy

treated with postoperative radiotherapy. Int J Radiat Oncol Biol Phys 1999; 43:

1049–1053.

25. Huerta Bahena J, Labastida Almendaro S, Cortez Arroyo H, Calva A.

[Postoperative radiotherapy in patients with invasive uterine cervix cancer treated

previously with simple hysterectomy. Results from the Hospital de Oncologia,

Centro Medico Nacional SXXI.]. Ginecol Obstet Mex 2003; 71: 304–311.

26. Chen SW, Liang JA, Yang SN, Lin FJ. Postoperative radiotherapy for patients with

invasive cervical cancer following treatment with simple hysterectomy. Jpn J Clin

Oncol 2003; 33: 477–481.

27. Durrance FY. Radiotherapy following simple hysterectomy in patients with stage I

and II carcinoma of the cervix. Am J Roentgenol Radium Ther Nucl Med 1968;

102: 165–169.

28. Hamberger AD, Unal A, Gershenson DM, Fletcher GH. Analysis of the severe

complications of irradiation of carcinoma of the cervix: whole pelvis irradiation

and intracavitary radium. Int J Radiat Oncol Biol Phys 1983; 9: 367–371.

29. Komaki R, Brickner TJ, Hanlon AL et al. Long-term results of treatment of

cervical carcinoma in the United States in 1973, 1978, and 1983: Patterns of

Care Study (PCS). Int J Radiat Oncol Biol Phys 1995; 31: 973–982.

30. Lanciano RM, Martz K, Montana GS, Hanks GE. Influence of age, prior abdominal

surgery, fraction size, and dose on complications after radiation therapy for

squamous cell cancer of the uterine cervix. A patterns of care study. Cancer

1992; 69: 2124–2130.

31. Thomas G, Dembo A, Fyles A et al. Concurrent chemoradiation in advanced

cervical cancer. Gynecol Oncol 1990; 38: 446–451.

32. Green TH Jr, Morse WJ Jr. Management of invasive cervical cancer following

inadvertent simple hysterectomy. Obstet Gynecol 1969; 33: 763–769.

33. Orr JW Jr, Ball GC, Soong SJ et al. Surgical treatment of women found to have

invasive cervix cancer at the time of total hysterectomy. Obstet Gynecol 1986;

68: 353–356.

34. Kinney WK, Egorshin EV, Ballard DJ, Podratz KC. Long-term survival and

sequelae after surgical management of invasive cervical carcinoma diagnosed at

the time of simple hysterectomy. Gynecol Oncol 1992; 44: 24–27.

35. Chapman JA, Mannel RS, DiSaia PJ et al. Surgical treatment of unexpected

invasive cervical cancer found at total hysterectomy. Obstet Gynecol 1992; 80:

931–934.

original article Annals of Oncology

1000 | Park et al. Volume 21 | No. 5 | May 2010

by guest on October 4, 2015

http://annonc.oxfordjournals.org/D

ownloaded from