Preoperative Serum Squamous Cell Carcinoma Antigen Levels in Clinical Decision Making for Patients With Early-Stage Cervical Cancer
http://www.100md.com
《临床肿瘤学》
the Departments of Gynecological Oncology, Pathology, Radiotherapy, Medical Oncology, and Epidemiology and Statistics, University Hospital Groningen
Department of Gynecological Oncology, University Hospital Amsterdam, Amsterdam
Department of Obstetrics and Gynecology, Meander Medical Centre, Amersfoort, The Netherlands.
ABSTRACT
PATIENTS AND METHODS: In a cohort study, International Federation of Gynecology and Obstetrics (FIGO) stage, tumor size, and preoperative serum SCC-ag levels, as determined by enzyme immunoassay, were related to the frequency of postoperative indications for adjuvant radiotherapy in 337 surgically treated, FIGO stage IB/IIA, squamous cell cervical cancer patients.
RESULTS: In patients with normal preoperative SCC-ag, 16% of IB1 and 29% of IB2/IIA had postoperative indications for adjuvant radiotherapy, in contrast to 57% of IB1 and 74% of IB2/IIA patients with elevated (> 1.9 ng/mL) serum SCC-ag (P < .001). Serum SCC-ag was the only independent predictor for a postoperative indication for radiotherapy (odds ratio, 7.1; P < .001). Furthermore, in IB1 patients that did not have indications for adjuvant radiotherapy, 15% of patients with elevated preoperative serum SCC-ag levels recurred within 2 years, compared with 1.6% of patients with normal serum SCC-ag levels (P = .02).
CONCLUSION: In early-stage cervical cancer, determination of serum SCC-ag levels allows more refined preoperative estimation of the likelihood for adjuvant radiotherapy than current clinical parameters, and simultaneously identifies patients at high risk for recurrence when treated with surgery only. The role of preoperative serum SCC-ag in the management of patients with early-stage cervical cancer deserves further investigation.
INTRODUCTION
The National Institutes of Health consensus statement of 1997 on cervical cancer reported that radical hysterectomy (combined with adjuvant radiotherapy [RT] when indicated) and primary RT are equally effective treatment modalities for early-stage (FIGO IB/IIA) cervical cancer.3 This consensus opinion was later confirmed by a randomized trial.4 However, controversy on treatment choices in early-stage cervical cancer still remains, and has been the genesis of, protocol GOG (Gynecologic Oncology Group) -0201. In this randomized trial, radical hysterectomy, followed by tailored chemoradiation, is compared with primary chemoradiation in patients with stage IB2 cervical cancer. Progression-free survival, treatment-related toxicity, and health-related quality of life are the end points in this important study.
Squamous cell cancer antigen (SCC-ag) is a known serum tumor marker for squamous cell cervical cancer. Most studies reported that elevated preoperative serum SCC-ag is associated with pelvic lymph-node metastases, but independently from lymph-node metastases with worse survival.5-11 All of these (often too small) studies used different single cutoffs for preoperative serum SCC-ag analysis and analyzed predictive values for lymph-node metastases only, which is just one of the three classic indications for radiotherapy. Due to these limitations, until now, not enough nuances were available to allow widespread implementation of pretreatment serum SCC-ag analysis in clinical decision making.
To obtain a better informed choice between primary surgery and primary chemoradiation in early-stage squamous cell cervical cancer patients, in the present study, we analyzed the predictive value of a continuum of preoperative serum SCC-ag levels for either one or a combination of postoperative indications for adjuvant RT and compared the performance of serum SCC-ag analysis to currently used clinical parameters, that is, FIGO stage and tumor size.
PATIENTS AND METHODS
Staging, Treatment, and Follow-Up
Pelvic examination under general anesthesia was performed for staging in accordance with the FIGO criteria.2 During this procedure, lesion size (largest diameter) and tumor spread beyond the cervix were estimated routinely. Tumor size was expressed in centimeters.
Patients with FIGO stages IB and IIA were eligible for surgical treatment which consisted of a class III radical hysterectomy with pelvic lymphadenectomy in both institutions.14 Exceptions were made for patients with poor general conditions for whom surgery determined an unacceptable risk. When primary radical surgery was abandoned during operation due to obvious extracervical tumor mass, patients were treated with primary RT, which was combined with chemotherapy (three cycles of carboplatin 300 mg/m2 on day 1 and fluorouracil 600 mg/m2 on days 2 to 5, and from January 1, 2002, cisplatin 40 mg/m2/week for 6 weeks) at the University Hospital Groningen.
In both hospitals, patients received adjuvant RT when considered at high risk for locoregional recurrence. Classic high-risk indications were: one or more pelvic lymph node metastases; close (< 2 mm) or positive resection margins; and parametrial involvement. Patients considered to be at intermediate risk (with either deep stromal infiltration [more than two thirds of the cervical stroma] and/or massive capillary/lymphatic space involvement [at least three different tumor localizations in capillary or lymphatic space to be seen outside of the main tumor mass in at least two slides using a 10 x 20 enlargement]) only received adjuvant RT at the University Hospital Groningen. Adjuvant RT consisted of external RT (6 to 10 MV) to the pelvis, using the box-technique, five fractions a week of 1.8 Gy, to a total dose of 45 Gy.
Follow-Up
Standard follow-up surveillance consisted of a clinical history, physical examination, and blood analysis starting 6 weeks after treatment, then every 2 months for the first year, every 3 months in the second year, then every 6 months until the sixth year, and yearly thereafter until 10 years after treatment.
Institutional Review Board Approval
All clinicopathologic, follow-up, and serum SCC-ag data were obtained during standard treatment and follow-up. For the present study, all relevant data were retrieved from our previously mentioned, larger, computerized database into a separate anonymous database. In this separate database, password protected patient identity was protected by study-specific, unique patient codes, which were only known to two dedicated data managers, who also have daily responsibility for the larger database. In case of uncertainties with respect to clinicopathologic, follow-up, and serum SCC-ag data, the larger databases could only be checked through the data managers, thereby ascertaining the protection of patients’ identity. Based on this information, our local ethics committee/institutional review board decided that for this study, no further approval was needed.
SCC-ag Analysis
During the study period, it was clinical practice in both hospitals to use serum SCC-ag analysis for the confirmation of complete remission after surgery and for the early detection of recurrences. To serve as a reference, a serum sample was routinely taken at the patient’s initial visit. SCC-ag was measured by the IMx SCC-ag microparticle enzyme immunoassay (Abbott Laboratories, Abbott Park, IL). Treatment decisions were taken independently of the preoperative SCC-ag value. The standard cutoff value used for serum SCC-ag was > 1.9 ng/mL, which is the 99th percentile in a population of 250 healthy women tested at the University Hospital Groningen.
Statistical Analysis
Differences in the distributions of age, stage, and other patient characteristics were analyzed with the {chi}2 test. The ability to preoperatively identify those patients with a classic indication for adjuvant RT by serum SCC-ag analysis was compared with the performance of tumor size and FIGO stage by calculating receiver operating characteristic (ROC) curves. Multivariate analysis was performed using the Cox regression model, entering serum SCC-ag, tumor size, and FIGO stage into the model in a one-step analysis. For the ROC curves and regression analysis, one cutoff level was chosen for the three variables used: SCC-ag = 1.9 ng/mL; tumor size > 4 cm, and stage IB versus IIA. For evaluation of serum SCC-ag analysis as a diagnostic tool to preoperatively identify those patients who will receive adjuvant RT for classic indications, positive and negative predictive values were calculated for the SCC-ag cutoff values from 1.0 through 10.0, starting at 1.0 and recalculating cutoff values until 10.0, at a rate increasing by 1.0. The association between increasing levels of SCC-ag and the chance of receiving adjuvant RT was evaluated with the {chi}2 test for trend. As most locoregional recurrences of early-stage cervical cancer occur within 2 years after primary treatment,15,16 analysis of 2-year recurrence percentages was performed using the {chi}2 test or Fisher’s exact test when appropriate. Differences associated with a P value ≤ .05 were considered significant.
All of these analyses were performed using SPSS software (version 11, SPSS Inc, Chicago, IL). Ninety-five percent confidence intervals for proportions of patients with indications for adjuvant RT were calculated using the CIA software (Gardner & British Medical Journal, London, United Kingdom).
RESULTS
Study Group Characteristics
Median age at time of diagnosis was 41 years (interquartile-range [IQR], 34 to 50). FIGO stage IB cervical cancer was diagnosed in 276 (82%) patients (198 [72%] stage IB1; 61 [18%] stage IB2) and FIGO stage IIA in 61 (18%) patients. Adjuvant RT was given for classic indications in 110 (33%) patients (presence of pelvic lymph node metastases in 80 [73%], parametrial involvement in 16 [15%], tumor positive resection margins in eight [7.3%], and a combination of these three in six [5.5%] patients).
In 18 patients macroscopic extracervical disease (confirmed by frozen section) was found during surgery, after which the radical hysterectomy was abandoned and the patients were treated with primary chemoradiation. For the evaluation of the value of serum SCC-ag to predict classic indications for adjuvant RT, these 18 patients were categorized as having classic indications for adjuvant RT, as these patients finally received both surgery and RT (combined with chemotherapy in four patients). These 18 patients were not included for the evaluation of recurrence percentages in patients treated with radical hysterectomy and adjuvant RT.
Only nine patients received adjuvant RT because of intermediate risk. Therefore, it was decided not to analyze the predictive value of serum SCC-ag levels for intermediate-risk separately. For the evaluation of the value of serum SCC-ag to predict classic indications for adjuvant RT, these nine patients were classified as having no classic indication for adjuvant RT. The data of these nine patients were not used in the analyses of recurrence percentages in patients only treated with radical hysterectomy.
The overall frequency of indications for adjuvant RT, including the 27 patients who were reclassified, is shown in Table 1. In stage IB2 and IIA patients, a comparable frequency of indications for adjuvant RT and of elevated serum SCC-ag was observed and therefore both groups were combined for further analysis. There was no difference in the distribution of age, stage, frequency of adjuvant RT for classic indications, or frequency of recurrence between the two hospitals.
FIGO Stage, Tumor Size, and Serum SCC-ag in Relation to Indications for Adjuvant RT
Preoperative serum SCC-ag levels were elevated (> 1.9 ng/mL) in 128 (38%) patients (Table 1). Table 2 represents the frequency of elevated serum SCC-ag levels for patients without any indication for adjuvant RT and for the different indications for RT separately. The frequency of elevated serum SCC-ag was highest in patients with more than one classic indication for adjuvant RT. Figure 1, composed of the ROC curves, represents the identification of patients with a classic indication for adjuvant RT by the FIGO stage, tumor size, and serum SCC-ag criteria, respectively. The better performance as a predictor of classic indications for adjuvant RT of serum SCC-ag compared with tumor size and stage is expressed by its ROC curve, which is closest to the upper left corner of the graph. The sensitivity and specificity to predict classic indications for adjuvant therapy was 27% and 87% for FIGO stage, 47% and 79% for tumor size > 4 cm, and 68% and 80% for SCC-ag level > 1.9 ng/mL. Elevated serum SCC-ag was the only independent predictor for indications for adjuvant RT with an odds ratio of 7.1 in multivariate analysis including tumor size and FIGO stage (Table 3).
Figure 2 shows the frequency of adjuvant RT because of classic indications for increasing levels of preoperative serum SCC-ag. The higher the level of preoperative serum SCC-ag, the higher the frequency of adjuvant RT (P < .001). Using our classic cutoff value (1.9 ng/mL), serum SCC-ag analysis allowed the identification of a group of patients at high risk for adjuvant RT in both stage IB1 and stage IB2/IIA patients. In 27 (57%) of 47 stage IB1 patients with elevated serum SCC-ag, classic indications for adjuvant RT were present, compared with 24 (16%) of 151 stage IB1 patients with normal preoperative serum SCC-ag (P < .001). In 60 (74%) of 81 stage IB2/IIA patients with elevated serum SCC-ag, classic indications for adjuvant RT were present, compared with 17 (29%) of 58 stage IB2/IIA patients with normal preoperative serum SCC-ag (P < .001).
The predictive value of serum SCC-ag for classic indications for adjuvant RT is shown in Figure 3 for a continuum of different cutoff values. This figure allows evaluation of the implications of choosing a single cutoff value at different levels for serum SCC-ag for all early-stage cervical cancer patients. When a cutoff value of 2.0 ng/mL is chosen, the likelihood of having an indication for adjuvant RT in the patients treated with surgery is approximately 16% in stage IB1 and 22% in stage IB2/IIA patients. Choosing a higher cutoff, for example, 4.0 ng/mL, will lead to higher positive predictive values. When preoperative serum SCC-ag is higher than 4.0 ng/mL, the likelihood of having an indication for adjuvant RT will be approximately 80% for both stage IB1 and stage IB2/IIA patients.
All analyses were also performed for both hospitals individually. No differences were observed between the data of the two hospitals separately.
The Association of Elevated Serum SCC-ag and Recurrence of Disease
The median follow-up time in our population of 337 early-stage cervical cancer patients was 3.9 years (IQR, 1.4 to 6.2). Twenty-one patients were lost to follow-up for a variety of reasons after a median follow-up time of 3.3 years (IQR, 0.9 to 5.4). Overall, 54 recurrences were observed and the median time to recurrence was 1.0 year (IQR, 0.6 to 1.8; range, 0.3 to 10.3 years). Of these 54 recurrences, 22 occurred in patients with normal preoperative serum SCC-ag and 32 in patients with an elevated level. At the time of recurrence, nine patients (41%) with normal preoperative serum SCC-ag and 15 patients (47%) with preoperatively elevated levels had distant metastases.
In the group of 198 patients with stage IB1 disease, 144 were not treated with adjuvant RT. In patients with normal preoperative serum SCC-ag, two (1.6%) of 124 had a recurrence within 2 years, while in the remaining 20 patients with elevated serum SCC-ag, three patients (15%) had recurrent disease (P = .02; Fig 4). In 49 stage IB1 patients that received adjuvant RT for classic indications, no differences in frequency of recurrences were observed between patients with normal (four recurrences in 23 patients) and elevated preoperative serum SCC-ag (four recurrences in 26 patients).
In stage IB2/IIA patients, serum SCC-ag levels did not identify patients at higher risk for recurrence, neither in patients only treated with surgery nor in patients treated with surgery and adjuvant RT.
DISCUSSION
Patient characteristics of our study population are comparable to the characteristics of other early-stage cervical cancer populations previously described.2,4,17 The relative high percentage of adjuvant RT in our study can be explained by the exclusion of 184 patients with previous loop excision or exconisation, as these patients have particularly small tumors at low risk for adjuvant RT.
RT as primary treatment can be combined with chemotherapy because recent trials (primarily in more advanced stages of disease) show a survival benefit for this combined modality treatment in cervical cancer.18 While a role for surgery has been clearly established in patients with stage IB1 cervical cancer, currently available evidence has resulted in a considerable controversy on the role of surgery in bulky cervical tumors.2,19-21 Those favoring surgery cite its better identification of extent of disease (occult para-aortic nodal disease, occult intraperitoneal spread), removal of involved nodes, its better preservation of ovarian and sexual functioning and debulking of large tumor volumes. Arguments against surgery include the increased morbidity of two-modality treatment, the higher costs of two rather than one therapy modality, and the current lack of clear benefit in terms of survival of radical hysterectomy plus/minus chemoradiation therapy versus primary chemoradiation alone. Based on the previously mentioned considerations, more surgically oriented gynecologic oncologists often favor surgery followed by tailored chemoradiation for all early-stage patients, while in contrast, others state that two-modality treatment should be avoided as much as possible.3,19,20,22 In some institutions, radical surgery has been restricted to stage IB1 patients with tumors ≤ 3 cm, thus reducing the frequency of adjuvant RT to approximately 12% of the operated patients.23 In editorials on treatment choices in early-stage cervical cancer, Grigsby19 and Morris20 both suggested that criteria for selecting patients for radical hysterectomy should be defined such that postoperative adjuvant RT will be given to a maximum of 5% to 20% of the patients. To end this controversy on optimal treament for early-stage cervical cancer, a randomized trial is currently performed by the GOG in stage IB2 patients comparing primary chemoradiation versus primary radical hysterectomy followed by tailored chemoradiation (protocol GOG-0201). While awaiting the results of this trial, preoperative serum SCC-ag analysis may allow gynecologic oncologists and their patients a more precise estimation of how high their chances are for two-modality treatment. When this more precise estimation has been confirmed, one also may consider treating patients with a higher likelihood for adjuvant RT with primary RT in order to avoid the increased morbidity of two-modality treatment in a significant proportion of patients. Our study suggests that when knowledge of FIGO stage and SCC-ag levels are combined, the group of patients with FIGO stage IB1 and preoperative serum SCC-ag levels between 0 and 1.5 ng/mL (Fig 2) have a low likelihood of two-modality therapy. When in stage IB1 patients, the classic cutoff for SCC-ag of 1.9 ng/mL is used, approximately 16% of the patients will ultimately have an indication for adjuvant RT (Fig 4). Our study also indicates that in stage IB2/IIA, only patients with serum SCC-ag levels between 0 and 1.0 ng/mL have a frequency of adjuvant RT close to 10% to 20%. However, these data should be used with caution since only a small group of stage IB2/IIA patients (n = 31) had a preoperative serum SCC-ag value of 1.0 ng/mL or lower. Although thoroughly discussed, the choice to use a pretreatment likelihood of 5% to 20% for adjuvant RT as a criterion for eligibility for radical surgery remains an arbitrary one. An advantage of the data presented in Figures 2 and 3 is that it allows pretreatment serum SCC-ag analysis to be implemented in existing clinical guidelines because different SCC-ag levels corresponding to different likelihoods for two-modality treatment can be chosen as cutoffs in the choice between primary surgery and chemoradiation. Furthermore, based on the preoperative SCC-ag level, data as presented in Figure 2 may also be used to more precisely inform an individual patient of her chance for adjuvant chemoradiation after radical surgery.
An obvious drawback of possible implementation of preoperative serum SCC-ag as a cutoff in choice between primary surgery and primary RT is the treatment with primary RT for patients who, based on classic indications, would not have received adjuvant RT after surgery (eg, 43% of the stage IB1 patients with preoperative serum SCC-ag > 1.9 ng/mL; Fig 4), and one could argue that treatment by primary RT instead of radical hysterectomy only may cause unnecessary higher long-term morbidity in these patients. Again, the data of the ongoing GOG-0201 will shed more light on these issues. However, our data suggest that especially those stage IB1 patients with elevated serum SCC-ag are at higher risk for recurrence when treated with radical surgery only. In 15% of stage IB1 patients with elevated serum SCC-ag and no classic indication for adjuvant RT, recurrent disease was observed compared with 1.6% recurrent disease in otherwise comparable stage IB1 patients with normal serum SCC-ag. This observation indicates that preoperative serum SCC-ag analysis also allows identification of patients at greater risk for recurrence after radical surgery who would not have been identified otherwise. This patient group may particularly benefit from choosing chemoradiation as primary treatment, as prognosis after a recurrence is poor, with most patients dying as a result of uncontrolled disease.20 Another randomized trial will be the only way to prove the benefit of chemoradiation over radical hysterectomy in stage IB1 patients with elevated serum SCC-ag levels.
Until now, previous studies on preoperative serum SCC-ag levels have not led to a wide implementation of preoperative serum SCC-ag analysis into clinical practice, due to small study populations, the lack of including parametrial involvement and positive resection margins into a predictive model, and/or inclusion of more advanced-stage patients into the predictive model. Despite the restrictions mentioned, these studies also showed strong associations between elevated preoperative serum SCC-ag levels and the presence of lymph node metastases, thus underlining the strength of the data reported in our present study.6,8,10,11,24 The biggest hurdle preventing implementation of serum SCC-ag analysis in clinical decision making is probably the previously mentioned controversy on best primary treatment choice for early-stage cervical cancer. Those authorities in favor of primary chemoradiation in patients at high risk for adjuvant treatment after surgery will welcome any evidence on a more refined risk assessment by serum SCC-ag analysis. On the other hand, those authorities in favor of surgery will not change their point of view by reading our article, since we do not provide new evidence on which therapeutic strategy is best. However, we do provide data that show that SCC-ag analysis may be a tool (different cutoff values possible) by which multimodality treatment can be predicted by evidence-based means.
In both of our institutions, imaging techniques, such as magnetic resonance imaging or computed tomography, are not routinely used in staging patients with cervical cancer, although it has been reported that these imaging techniques can be used for either staging of cervical cancer or for preoperative identification of lymph-node metastases.25 However, it has also been shown that interobserver correlations are rather low, with {kappa} values between 0.28 and 0.62, and that the quality of images largely depend on the equipment and different techniques used.21,26,27 In contrast, serum SCC-ag analysis can easily be performed everywhere with a relatively cheap, commercially available enzyme immunoassay, which is not subject to interobserver variation. Serum SCC-ag analysis can be reliably used by every oncologist, and therefore our results may be used to identify future squamous cell cervical cancer patients at different levels of risk for adjuvant RT after surgery (Fig 2).
Another possible application of preoperative serum SCC-ag analysis is stratification of early-stage cervical cancer patients in different risk groups when designing clinical trials. In patients with early-stage squamous cell cervical cancer, clinical trials can, apart from tumor size and FIGO stage, stratify for patients with elevated serum SCC-ag levels because of the clearly different risk profile. Studies on more intensive treatment regimens (such as combining chemotherapy and RT) can then be restricted to early-stage cervical cancer patients at high risk for recurrence, preventing additive morbidity in patients that are adequately treated with surgery.
In conclusion, our current study shows that preoperative serum SCC-ag analysis may allow more refined and individualized preoperative risk assessment for adjuvant RT than by currently used clinical parameters. After confirmation of our data in independent study populations, further implementation of pretreatment serum SCC-ag analysis in clinical decision making will be largely dependent on the outcome of the ongoing debate on the best choice of primary treatment for early-stage cervical cancer patients.
Authors’ Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Dr Gillian M. Thomas for critically reading and reviewing our manuscript.
NOTES
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Parkin DM, Pisani P, Ferlay J: Global cancer statistics. CA Cancer J Clin 49:33-64,1, 1999
2. Finan MA, DeCesare S, Fiorica JV, et al: Radical hysterectomy for stage IB1 vs IB2 carcinoma of the cervix: Does the new staging system predict morbidity and survival Gynecol Oncol 62:139-147, 1996
3. National Institutes of Health Consensus Development Conference statement on cervical cancer. April 1-3, 1996. Gynecol Oncol 66:351-361, 1997
4. Landoni F, Maneo A, Colombo A, et al: Randomised study of radical surgery versus radiotherapy for stage Ib- IIa cervical cancer. Lancet 350:535-540, 1997
5. Duk JM, Groenier KH, de Bruijn HW, et al: Pretreatment serum squamous cell carcinoma antigen: A newly identified prognostic factor in early-stage cervical carcinoma. J Clin Oncol 14:111-118, 1996
6. Bolger BS, Dabbas M, Lopes A, et al: Prognostic value of preoperative squamous cell carcinoma antigen level in patients surgically treated for cervical carcinoma. Gynecol Oncol 65:309-313, 1997
7. Juang CM, Wang PH, Yen MS, et al: Application of tumor markers CEA, TPA, and SCC-Ag in patients with low-risk FIGO stage IB and IIA squamous cell carcinoma of the uterine cervix. Gynecol Oncol 76:103-106, 2000
8. Strauss HG, Laban C, Lautenschlager C, et al: SCC antigen in the serum as an independent prognostic factor in operable squamous cell carcinoma of the cervix. Eur J Cancer 38:1987-1991, 2002
9. Gaarenstroom KN, Kenter GG, Bonfrer JM, et al: Can initial serum cyfra 21-1, SCC antigen, and TPA levels in squamous cell cervical cancer predict lymph node metastases or prognosis Gynecol Oncol 77:164-170, 2000
10. Lin H, ChangChien CC, Huang EY, et al: The role of pretreatment squamous cell carcinoma antigen in predicting nodal metastasis in early stage cervical cancer. Acta Obstet Gynecol Scand 79:140-144, 2000
11. Takeshima N, Hirai Y, Katase K, et al: The value of squamous cell carcinoma antigen as a predictor of nodal metastasis in cervical cancer. Gynecol Oncol 68:263-266, 1998
12. Kurman RJ, Norris HJ, Wilkinson E: Tumors of the cervix, vagina, and vulva: Tumors of the cervix, in Rosai J (ed): Atlas of Tumor Pathology. Washington, DC, Armed Forces Institute of Pathology, 1992 (pp 37-139)
13. Maruo T, Shibata K, Kimura A, et al: Tumor-associated antigen, TA-4, in the monitoring of the effects of therapy for squamous cell carcinoma of the uterine cervix. Serial determinations and tissue localization. Cancer 56:302-308, 1985
14. Hacker NF: Cervical cancer, in Berek JS, Hacker NF (eds): Practical Gynecologic Oncology. Philadelphia, PA, Lippincott Williams & Wilkins, 2000 (345-407)
15. Friedlander M: Guidelines for the treatment of recurrent and metastatic cervical cancer. Oncologist 7:342-347, 2002
16. Look KY, Rocereto TF: Relapse patterns in FIGO stage IB carcinoma of the cervix. Gynecol Oncol 38:114-120, 1990
17. Sedlis A, Bundy BN, Rotman MZ, et al: A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: A Gynecologic Oncology Group Study. Gynecol Oncol 73:177-183, 1999
18. de Roda H, Walboomers JM, Meijer CJ, et al: Analysis of cytomorphologically abnormal cervical scrapes for the presence of 27 mucosotropic human papillomavirus genotypes, using polymerase chain reaction. Int J Cancer 56:802-806, 1994
19. Grigsby PW: Stage IB1 vs IB2 carcinoma of the cervix: Should the new FIGO staging system define therapy Gynecol Oncol 62:135-136, 1996
20. Morris M: Early cervical carcinoma: Are two treatments better than one Gynecol Oncol 54:1-3, 1994
21. Bloss JD, Berman ML, Mukhererjee J, et al: Bulky stage IB cervical carcinoma managed by primary radical hysterectomy followed by tailored radiotherapy. Gynecol Oncol 47:21-27, 1992
22. Burke TW: Treatment options in stage IB cervical cancer: Radical hysterectomy and radiotherapy. Semin Radiat Oncol 4:34-40, 1994
23. Covens A, Rosen B, Murphy J, et al: Changes in the demographics and perioperative care of stage IA(2)/IB(1) cervical cancer over the past 16 years. Gynecol Oncol 81:133-137, 2001
24. Chou CY, Wang ST, Kuo HC, et al: Serum level of squamous cell carcinoma antigen and tumor size are useful to identify preoperatively patients at high risk of cervical cancer. Cancer 74:2497-2501, 1994
25. Boss EA, Barentsz JO, Massuger LF, et al: The role of MR imaging in invasive cervical carcinoma. Eur Radiol 10:256-270, 2000
26. Guyatt GH, Lefcoe M, Walter S, et al: Interobserver variation in the computed tomographic evaluation of mediastinal lymph node size in patients with potentially resectable lung cancer. Canadian Lung Oncology Group. Chest 107:116-119, 1995
27. Stafira JS, Sonnad JR, Yuh WT, et al: Qualitative assessment of cervical spinal stenosis: Observer variability on CT and MR images. AJNR Am J Neuroradiol 24:766-769, 2003(Nathalie Reesink-Peters, )
Department of Gynecological Oncology, University Hospital Amsterdam, Amsterdam
Department of Obstetrics and Gynecology, Meander Medical Centre, Amersfoort, The Netherlands.
ABSTRACT
PATIENTS AND METHODS: In a cohort study, International Federation of Gynecology and Obstetrics (FIGO) stage, tumor size, and preoperative serum SCC-ag levels, as determined by enzyme immunoassay, were related to the frequency of postoperative indications for adjuvant radiotherapy in 337 surgically treated, FIGO stage IB/IIA, squamous cell cervical cancer patients.
RESULTS: In patients with normal preoperative SCC-ag, 16% of IB1 and 29% of IB2/IIA had postoperative indications for adjuvant radiotherapy, in contrast to 57% of IB1 and 74% of IB2/IIA patients with elevated (> 1.9 ng/mL) serum SCC-ag (P < .001). Serum SCC-ag was the only independent predictor for a postoperative indication for radiotherapy (odds ratio, 7.1; P < .001). Furthermore, in IB1 patients that did not have indications for adjuvant radiotherapy, 15% of patients with elevated preoperative serum SCC-ag levels recurred within 2 years, compared with 1.6% of patients with normal serum SCC-ag levels (P = .02).
CONCLUSION: In early-stage cervical cancer, determination of serum SCC-ag levels allows more refined preoperative estimation of the likelihood for adjuvant radiotherapy than current clinical parameters, and simultaneously identifies patients at high risk for recurrence when treated with surgery only. The role of preoperative serum SCC-ag in the management of patients with early-stage cervical cancer deserves further investigation.
INTRODUCTION
The National Institutes of Health consensus statement of 1997 on cervical cancer reported that radical hysterectomy (combined with adjuvant radiotherapy [RT] when indicated) and primary RT are equally effective treatment modalities for early-stage (FIGO IB/IIA) cervical cancer.3 This consensus opinion was later confirmed by a randomized trial.4 However, controversy on treatment choices in early-stage cervical cancer still remains, and has been the genesis of, protocol GOG (Gynecologic Oncology Group) -0201. In this randomized trial, radical hysterectomy, followed by tailored chemoradiation, is compared with primary chemoradiation in patients with stage IB2 cervical cancer. Progression-free survival, treatment-related toxicity, and health-related quality of life are the end points in this important study.
Squamous cell cancer antigen (SCC-ag) is a known serum tumor marker for squamous cell cervical cancer. Most studies reported that elevated preoperative serum SCC-ag is associated with pelvic lymph-node metastases, but independently from lymph-node metastases with worse survival.5-11 All of these (often too small) studies used different single cutoffs for preoperative serum SCC-ag analysis and analyzed predictive values for lymph-node metastases only, which is just one of the three classic indications for radiotherapy. Due to these limitations, until now, not enough nuances were available to allow widespread implementation of pretreatment serum SCC-ag analysis in clinical decision making.
To obtain a better informed choice between primary surgery and primary chemoradiation in early-stage squamous cell cervical cancer patients, in the present study, we analyzed the predictive value of a continuum of preoperative serum SCC-ag levels for either one or a combination of postoperative indications for adjuvant RT and compared the performance of serum SCC-ag analysis to currently used clinical parameters, that is, FIGO stage and tumor size.
PATIENTS AND METHODS
Staging, Treatment, and Follow-Up
Pelvic examination under general anesthesia was performed for staging in accordance with the FIGO criteria.2 During this procedure, lesion size (largest diameter) and tumor spread beyond the cervix were estimated routinely. Tumor size was expressed in centimeters.
Patients with FIGO stages IB and IIA were eligible for surgical treatment which consisted of a class III radical hysterectomy with pelvic lymphadenectomy in both institutions.14 Exceptions were made for patients with poor general conditions for whom surgery determined an unacceptable risk. When primary radical surgery was abandoned during operation due to obvious extracervical tumor mass, patients were treated with primary RT, which was combined with chemotherapy (three cycles of carboplatin 300 mg/m2 on day 1 and fluorouracil 600 mg/m2 on days 2 to 5, and from January 1, 2002, cisplatin 40 mg/m2/week for 6 weeks) at the University Hospital Groningen.
In both hospitals, patients received adjuvant RT when considered at high risk for locoregional recurrence. Classic high-risk indications were: one or more pelvic lymph node metastases; close (< 2 mm) or positive resection margins; and parametrial involvement. Patients considered to be at intermediate risk (with either deep stromal infiltration [more than two thirds of the cervical stroma] and/or massive capillary/lymphatic space involvement [at least three different tumor localizations in capillary or lymphatic space to be seen outside of the main tumor mass in at least two slides using a 10 x 20 enlargement]) only received adjuvant RT at the University Hospital Groningen. Adjuvant RT consisted of external RT (6 to 10 MV) to the pelvis, using the box-technique, five fractions a week of 1.8 Gy, to a total dose of 45 Gy.
Follow-Up
Standard follow-up surveillance consisted of a clinical history, physical examination, and blood analysis starting 6 weeks after treatment, then every 2 months for the first year, every 3 months in the second year, then every 6 months until the sixth year, and yearly thereafter until 10 years after treatment.
Institutional Review Board Approval
All clinicopathologic, follow-up, and serum SCC-ag data were obtained during standard treatment and follow-up. For the present study, all relevant data were retrieved from our previously mentioned, larger, computerized database into a separate anonymous database. In this separate database, password protected patient identity was protected by study-specific, unique patient codes, which were only known to two dedicated data managers, who also have daily responsibility for the larger database. In case of uncertainties with respect to clinicopathologic, follow-up, and serum SCC-ag data, the larger databases could only be checked through the data managers, thereby ascertaining the protection of patients’ identity. Based on this information, our local ethics committee/institutional review board decided that for this study, no further approval was needed.
SCC-ag Analysis
During the study period, it was clinical practice in both hospitals to use serum SCC-ag analysis for the confirmation of complete remission after surgery and for the early detection of recurrences. To serve as a reference, a serum sample was routinely taken at the patient’s initial visit. SCC-ag was measured by the IMx SCC-ag microparticle enzyme immunoassay (Abbott Laboratories, Abbott Park, IL). Treatment decisions were taken independently of the preoperative SCC-ag value. The standard cutoff value used for serum SCC-ag was > 1.9 ng/mL, which is the 99th percentile in a population of 250 healthy women tested at the University Hospital Groningen.
Statistical Analysis
Differences in the distributions of age, stage, and other patient characteristics were analyzed with the {chi}2 test. The ability to preoperatively identify those patients with a classic indication for adjuvant RT by serum SCC-ag analysis was compared with the performance of tumor size and FIGO stage by calculating receiver operating characteristic (ROC) curves. Multivariate analysis was performed using the Cox regression model, entering serum SCC-ag, tumor size, and FIGO stage into the model in a one-step analysis. For the ROC curves and regression analysis, one cutoff level was chosen for the three variables used: SCC-ag = 1.9 ng/mL; tumor size > 4 cm, and stage IB versus IIA. For evaluation of serum SCC-ag analysis as a diagnostic tool to preoperatively identify those patients who will receive adjuvant RT for classic indications, positive and negative predictive values were calculated for the SCC-ag cutoff values from 1.0 through 10.0, starting at 1.0 and recalculating cutoff values until 10.0, at a rate increasing by 1.0. The association between increasing levels of SCC-ag and the chance of receiving adjuvant RT was evaluated with the {chi}2 test for trend. As most locoregional recurrences of early-stage cervical cancer occur within 2 years after primary treatment,15,16 analysis of 2-year recurrence percentages was performed using the {chi}2 test or Fisher’s exact test when appropriate. Differences associated with a P value ≤ .05 were considered significant.
All of these analyses were performed using SPSS software (version 11, SPSS Inc, Chicago, IL). Ninety-five percent confidence intervals for proportions of patients with indications for adjuvant RT were calculated using the CIA software (Gardner & British Medical Journal, London, United Kingdom).
RESULTS
Study Group Characteristics
Median age at time of diagnosis was 41 years (interquartile-range [IQR], 34 to 50). FIGO stage IB cervical cancer was diagnosed in 276 (82%) patients (198 [72%] stage IB1; 61 [18%] stage IB2) and FIGO stage IIA in 61 (18%) patients. Adjuvant RT was given for classic indications in 110 (33%) patients (presence of pelvic lymph node metastases in 80 [73%], parametrial involvement in 16 [15%], tumor positive resection margins in eight [7.3%], and a combination of these three in six [5.5%] patients).
In 18 patients macroscopic extracervical disease (confirmed by frozen section) was found during surgery, after which the radical hysterectomy was abandoned and the patients were treated with primary chemoradiation. For the evaluation of the value of serum SCC-ag to predict classic indications for adjuvant RT, these 18 patients were categorized as having classic indications for adjuvant RT, as these patients finally received both surgery and RT (combined with chemotherapy in four patients). These 18 patients were not included for the evaluation of recurrence percentages in patients treated with radical hysterectomy and adjuvant RT.
Only nine patients received adjuvant RT because of intermediate risk. Therefore, it was decided not to analyze the predictive value of serum SCC-ag levels for intermediate-risk separately. For the evaluation of the value of serum SCC-ag to predict classic indications for adjuvant RT, these nine patients were classified as having no classic indication for adjuvant RT. The data of these nine patients were not used in the analyses of recurrence percentages in patients only treated with radical hysterectomy.
The overall frequency of indications for adjuvant RT, including the 27 patients who were reclassified, is shown in Table 1. In stage IB2 and IIA patients, a comparable frequency of indications for adjuvant RT and of elevated serum SCC-ag was observed and therefore both groups were combined for further analysis. There was no difference in the distribution of age, stage, frequency of adjuvant RT for classic indications, or frequency of recurrence between the two hospitals.
FIGO Stage, Tumor Size, and Serum SCC-ag in Relation to Indications for Adjuvant RT
Preoperative serum SCC-ag levels were elevated (> 1.9 ng/mL) in 128 (38%) patients (Table 1). Table 2 represents the frequency of elevated serum SCC-ag levels for patients without any indication for adjuvant RT and for the different indications for RT separately. The frequency of elevated serum SCC-ag was highest in patients with more than one classic indication for adjuvant RT. Figure 1, composed of the ROC curves, represents the identification of patients with a classic indication for adjuvant RT by the FIGO stage, tumor size, and serum SCC-ag criteria, respectively. The better performance as a predictor of classic indications for adjuvant RT of serum SCC-ag compared with tumor size and stage is expressed by its ROC curve, which is closest to the upper left corner of the graph. The sensitivity and specificity to predict classic indications for adjuvant therapy was 27% and 87% for FIGO stage, 47% and 79% for tumor size > 4 cm, and 68% and 80% for SCC-ag level > 1.9 ng/mL. Elevated serum SCC-ag was the only independent predictor for indications for adjuvant RT with an odds ratio of 7.1 in multivariate analysis including tumor size and FIGO stage (Table 3).
Figure 2 shows the frequency of adjuvant RT because of classic indications for increasing levels of preoperative serum SCC-ag. The higher the level of preoperative serum SCC-ag, the higher the frequency of adjuvant RT (P < .001). Using our classic cutoff value (1.9 ng/mL), serum SCC-ag analysis allowed the identification of a group of patients at high risk for adjuvant RT in both stage IB1 and stage IB2/IIA patients. In 27 (57%) of 47 stage IB1 patients with elevated serum SCC-ag, classic indications for adjuvant RT were present, compared with 24 (16%) of 151 stage IB1 patients with normal preoperative serum SCC-ag (P < .001). In 60 (74%) of 81 stage IB2/IIA patients with elevated serum SCC-ag, classic indications for adjuvant RT were present, compared with 17 (29%) of 58 stage IB2/IIA patients with normal preoperative serum SCC-ag (P < .001).
The predictive value of serum SCC-ag for classic indications for adjuvant RT is shown in Figure 3 for a continuum of different cutoff values. This figure allows evaluation of the implications of choosing a single cutoff value at different levels for serum SCC-ag for all early-stage cervical cancer patients. When a cutoff value of 2.0 ng/mL is chosen, the likelihood of having an indication for adjuvant RT in the patients treated with surgery is approximately 16% in stage IB1 and 22% in stage IB2/IIA patients. Choosing a higher cutoff, for example, 4.0 ng/mL, will lead to higher positive predictive values. When preoperative serum SCC-ag is higher than 4.0 ng/mL, the likelihood of having an indication for adjuvant RT will be approximately 80% for both stage IB1 and stage IB2/IIA patients.
All analyses were also performed for both hospitals individually. No differences were observed between the data of the two hospitals separately.
The Association of Elevated Serum SCC-ag and Recurrence of Disease
The median follow-up time in our population of 337 early-stage cervical cancer patients was 3.9 years (IQR, 1.4 to 6.2). Twenty-one patients were lost to follow-up for a variety of reasons after a median follow-up time of 3.3 years (IQR, 0.9 to 5.4). Overall, 54 recurrences were observed and the median time to recurrence was 1.0 year (IQR, 0.6 to 1.8; range, 0.3 to 10.3 years). Of these 54 recurrences, 22 occurred in patients with normal preoperative serum SCC-ag and 32 in patients with an elevated level. At the time of recurrence, nine patients (41%) with normal preoperative serum SCC-ag and 15 patients (47%) with preoperatively elevated levels had distant metastases.
In the group of 198 patients with stage IB1 disease, 144 were not treated with adjuvant RT. In patients with normal preoperative serum SCC-ag, two (1.6%) of 124 had a recurrence within 2 years, while in the remaining 20 patients with elevated serum SCC-ag, three patients (15%) had recurrent disease (P = .02; Fig 4). In 49 stage IB1 patients that received adjuvant RT for classic indications, no differences in frequency of recurrences were observed between patients with normal (four recurrences in 23 patients) and elevated preoperative serum SCC-ag (four recurrences in 26 patients).
In stage IB2/IIA patients, serum SCC-ag levels did not identify patients at higher risk for recurrence, neither in patients only treated with surgery nor in patients treated with surgery and adjuvant RT.
DISCUSSION
Patient characteristics of our study population are comparable to the characteristics of other early-stage cervical cancer populations previously described.2,4,17 The relative high percentage of adjuvant RT in our study can be explained by the exclusion of 184 patients with previous loop excision or exconisation, as these patients have particularly small tumors at low risk for adjuvant RT.
RT as primary treatment can be combined with chemotherapy because recent trials (primarily in more advanced stages of disease) show a survival benefit for this combined modality treatment in cervical cancer.18 While a role for surgery has been clearly established in patients with stage IB1 cervical cancer, currently available evidence has resulted in a considerable controversy on the role of surgery in bulky cervical tumors.2,19-21 Those favoring surgery cite its better identification of extent of disease (occult para-aortic nodal disease, occult intraperitoneal spread), removal of involved nodes, its better preservation of ovarian and sexual functioning and debulking of large tumor volumes. Arguments against surgery include the increased morbidity of two-modality treatment, the higher costs of two rather than one therapy modality, and the current lack of clear benefit in terms of survival of radical hysterectomy plus/minus chemoradiation therapy versus primary chemoradiation alone. Based on the previously mentioned considerations, more surgically oriented gynecologic oncologists often favor surgery followed by tailored chemoradiation for all early-stage patients, while in contrast, others state that two-modality treatment should be avoided as much as possible.3,19,20,22 In some institutions, radical surgery has been restricted to stage IB1 patients with tumors ≤ 3 cm, thus reducing the frequency of adjuvant RT to approximately 12% of the operated patients.23 In editorials on treatment choices in early-stage cervical cancer, Grigsby19 and Morris20 both suggested that criteria for selecting patients for radical hysterectomy should be defined such that postoperative adjuvant RT will be given to a maximum of 5% to 20% of the patients. To end this controversy on optimal treament for early-stage cervical cancer, a randomized trial is currently performed by the GOG in stage IB2 patients comparing primary chemoradiation versus primary radical hysterectomy followed by tailored chemoradiation (protocol GOG-0201). While awaiting the results of this trial, preoperative serum SCC-ag analysis may allow gynecologic oncologists and their patients a more precise estimation of how high their chances are for two-modality treatment. When this more precise estimation has been confirmed, one also may consider treating patients with a higher likelihood for adjuvant RT with primary RT in order to avoid the increased morbidity of two-modality treatment in a significant proportion of patients. Our study suggests that when knowledge of FIGO stage and SCC-ag levels are combined, the group of patients with FIGO stage IB1 and preoperative serum SCC-ag levels between 0 and 1.5 ng/mL (Fig 2) have a low likelihood of two-modality therapy. When in stage IB1 patients, the classic cutoff for SCC-ag of 1.9 ng/mL is used, approximately 16% of the patients will ultimately have an indication for adjuvant RT (Fig 4). Our study also indicates that in stage IB2/IIA, only patients with serum SCC-ag levels between 0 and 1.0 ng/mL have a frequency of adjuvant RT close to 10% to 20%. However, these data should be used with caution since only a small group of stage IB2/IIA patients (n = 31) had a preoperative serum SCC-ag value of 1.0 ng/mL or lower. Although thoroughly discussed, the choice to use a pretreatment likelihood of 5% to 20% for adjuvant RT as a criterion for eligibility for radical surgery remains an arbitrary one. An advantage of the data presented in Figures 2 and 3 is that it allows pretreatment serum SCC-ag analysis to be implemented in existing clinical guidelines because different SCC-ag levels corresponding to different likelihoods for two-modality treatment can be chosen as cutoffs in the choice between primary surgery and chemoradiation. Furthermore, based on the preoperative SCC-ag level, data as presented in Figure 2 may also be used to more precisely inform an individual patient of her chance for adjuvant chemoradiation after radical surgery.
An obvious drawback of possible implementation of preoperative serum SCC-ag as a cutoff in choice between primary surgery and primary RT is the treatment with primary RT for patients who, based on classic indications, would not have received adjuvant RT after surgery (eg, 43% of the stage IB1 patients with preoperative serum SCC-ag > 1.9 ng/mL; Fig 4), and one could argue that treatment by primary RT instead of radical hysterectomy only may cause unnecessary higher long-term morbidity in these patients. Again, the data of the ongoing GOG-0201 will shed more light on these issues. However, our data suggest that especially those stage IB1 patients with elevated serum SCC-ag are at higher risk for recurrence when treated with radical surgery only. In 15% of stage IB1 patients with elevated serum SCC-ag and no classic indication for adjuvant RT, recurrent disease was observed compared with 1.6% recurrent disease in otherwise comparable stage IB1 patients with normal serum SCC-ag. This observation indicates that preoperative serum SCC-ag analysis also allows identification of patients at greater risk for recurrence after radical surgery who would not have been identified otherwise. This patient group may particularly benefit from choosing chemoradiation as primary treatment, as prognosis after a recurrence is poor, with most patients dying as a result of uncontrolled disease.20 Another randomized trial will be the only way to prove the benefit of chemoradiation over radical hysterectomy in stage IB1 patients with elevated serum SCC-ag levels.
Until now, previous studies on preoperative serum SCC-ag levels have not led to a wide implementation of preoperative serum SCC-ag analysis into clinical practice, due to small study populations, the lack of including parametrial involvement and positive resection margins into a predictive model, and/or inclusion of more advanced-stage patients into the predictive model. Despite the restrictions mentioned, these studies also showed strong associations between elevated preoperative serum SCC-ag levels and the presence of lymph node metastases, thus underlining the strength of the data reported in our present study.6,8,10,11,24 The biggest hurdle preventing implementation of serum SCC-ag analysis in clinical decision making is probably the previously mentioned controversy on best primary treatment choice for early-stage cervical cancer. Those authorities in favor of primary chemoradiation in patients at high risk for adjuvant treatment after surgery will welcome any evidence on a more refined risk assessment by serum SCC-ag analysis. On the other hand, those authorities in favor of surgery will not change their point of view by reading our article, since we do not provide new evidence on which therapeutic strategy is best. However, we do provide data that show that SCC-ag analysis may be a tool (different cutoff values possible) by which multimodality treatment can be predicted by evidence-based means.
In both of our institutions, imaging techniques, such as magnetic resonance imaging or computed tomography, are not routinely used in staging patients with cervical cancer, although it has been reported that these imaging techniques can be used for either staging of cervical cancer or for preoperative identification of lymph-node metastases.25 However, it has also been shown that interobserver correlations are rather low, with {kappa} values between 0.28 and 0.62, and that the quality of images largely depend on the equipment and different techniques used.21,26,27 In contrast, serum SCC-ag analysis can easily be performed everywhere with a relatively cheap, commercially available enzyme immunoassay, which is not subject to interobserver variation. Serum SCC-ag analysis can be reliably used by every oncologist, and therefore our results may be used to identify future squamous cell cervical cancer patients at different levels of risk for adjuvant RT after surgery (Fig 2).
Another possible application of preoperative serum SCC-ag analysis is stratification of early-stage cervical cancer patients in different risk groups when designing clinical trials. In patients with early-stage squamous cell cervical cancer, clinical trials can, apart from tumor size and FIGO stage, stratify for patients with elevated serum SCC-ag levels because of the clearly different risk profile. Studies on more intensive treatment regimens (such as combining chemotherapy and RT) can then be restricted to early-stage cervical cancer patients at high risk for recurrence, preventing additive morbidity in patients that are adequately treated with surgery.
In conclusion, our current study shows that preoperative serum SCC-ag analysis may allow more refined and individualized preoperative risk assessment for adjuvant RT than by currently used clinical parameters. After confirmation of our data in independent study populations, further implementation of pretreatment serum SCC-ag analysis in clinical decision making will be largely dependent on the outcome of the ongoing debate on the best choice of primary treatment for early-stage cervical cancer patients.
Authors’ Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Dr Gillian M. Thomas for critically reading and reviewing our manuscript.
NOTES
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Parkin DM, Pisani P, Ferlay J: Global cancer statistics. CA Cancer J Clin 49:33-64,1, 1999
2. Finan MA, DeCesare S, Fiorica JV, et al: Radical hysterectomy for stage IB1 vs IB2 carcinoma of the cervix: Does the new staging system predict morbidity and survival Gynecol Oncol 62:139-147, 1996
3. National Institutes of Health Consensus Development Conference statement on cervical cancer. April 1-3, 1996. Gynecol Oncol 66:351-361, 1997
4. Landoni F, Maneo A, Colombo A, et al: Randomised study of radical surgery versus radiotherapy for stage Ib- IIa cervical cancer. Lancet 350:535-540, 1997
5. Duk JM, Groenier KH, de Bruijn HW, et al: Pretreatment serum squamous cell carcinoma antigen: A newly identified prognostic factor in early-stage cervical carcinoma. J Clin Oncol 14:111-118, 1996
6. Bolger BS, Dabbas M, Lopes A, et al: Prognostic value of preoperative squamous cell carcinoma antigen level in patients surgically treated for cervical carcinoma. Gynecol Oncol 65:309-313, 1997
7. Juang CM, Wang PH, Yen MS, et al: Application of tumor markers CEA, TPA, and SCC-Ag in patients with low-risk FIGO stage IB and IIA squamous cell carcinoma of the uterine cervix. Gynecol Oncol 76:103-106, 2000
8. Strauss HG, Laban C, Lautenschlager C, et al: SCC antigen in the serum as an independent prognostic factor in operable squamous cell carcinoma of the cervix. Eur J Cancer 38:1987-1991, 2002
9. Gaarenstroom KN, Kenter GG, Bonfrer JM, et al: Can initial serum cyfra 21-1, SCC antigen, and TPA levels in squamous cell cervical cancer predict lymph node metastases or prognosis Gynecol Oncol 77:164-170, 2000
10. Lin H, ChangChien CC, Huang EY, et al: The role of pretreatment squamous cell carcinoma antigen in predicting nodal metastasis in early stage cervical cancer. Acta Obstet Gynecol Scand 79:140-144, 2000
11. Takeshima N, Hirai Y, Katase K, et al: The value of squamous cell carcinoma antigen as a predictor of nodal metastasis in cervical cancer. Gynecol Oncol 68:263-266, 1998
12. Kurman RJ, Norris HJ, Wilkinson E: Tumors of the cervix, vagina, and vulva: Tumors of the cervix, in Rosai J (ed): Atlas of Tumor Pathology. Washington, DC, Armed Forces Institute of Pathology, 1992 (pp 37-139)
13. Maruo T, Shibata K, Kimura A, et al: Tumor-associated antigen, TA-4, in the monitoring of the effects of therapy for squamous cell carcinoma of the uterine cervix. Serial determinations and tissue localization. Cancer 56:302-308, 1985
14. Hacker NF: Cervical cancer, in Berek JS, Hacker NF (eds): Practical Gynecologic Oncology. Philadelphia, PA, Lippincott Williams & Wilkins, 2000 (345-407)
15. Friedlander M: Guidelines for the treatment of recurrent and metastatic cervical cancer. Oncologist 7:342-347, 2002
16. Look KY, Rocereto TF: Relapse patterns in FIGO stage IB carcinoma of the cervix. Gynecol Oncol 38:114-120, 1990
17. Sedlis A, Bundy BN, Rotman MZ, et al: A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: A Gynecologic Oncology Group Study. Gynecol Oncol 73:177-183, 1999
18. de Roda H, Walboomers JM, Meijer CJ, et al: Analysis of cytomorphologically abnormal cervical scrapes for the presence of 27 mucosotropic human papillomavirus genotypes, using polymerase chain reaction. Int J Cancer 56:802-806, 1994
19. Grigsby PW: Stage IB1 vs IB2 carcinoma of the cervix: Should the new FIGO staging system define therapy Gynecol Oncol 62:135-136, 1996
20. Morris M: Early cervical carcinoma: Are two treatments better than one Gynecol Oncol 54:1-3, 1994
21. Bloss JD, Berman ML, Mukhererjee J, et al: Bulky stage IB cervical carcinoma managed by primary radical hysterectomy followed by tailored radiotherapy. Gynecol Oncol 47:21-27, 1992
22. Burke TW: Treatment options in stage IB cervical cancer: Radical hysterectomy and radiotherapy. Semin Radiat Oncol 4:34-40, 1994
23. Covens A, Rosen B, Murphy J, et al: Changes in the demographics and perioperative care of stage IA(2)/IB(1) cervical cancer over the past 16 years. Gynecol Oncol 81:133-137, 2001
24. Chou CY, Wang ST, Kuo HC, et al: Serum level of squamous cell carcinoma antigen and tumor size are useful to identify preoperatively patients at high risk of cervical cancer. Cancer 74:2497-2501, 1994
25. Boss EA, Barentsz JO, Massuger LF, et al: The role of MR imaging in invasive cervical carcinoma. Eur Radiol 10:256-270, 2000
26. Guyatt GH, Lefcoe M, Walter S, et al: Interobserver variation in the computed tomographic evaluation of mediastinal lymph node size in patients with potentially resectable lung cancer. Canadian Lung Oncology Group. Chest 107:116-119, 1995
27. Stafira JS, Sonnad JR, Yuh WT, et al: Qualitative assessment of cervical spinal stenosis: Observer variability on CT and MR images. AJNR Am J Neuroradiol 24:766-769, 2003(Nathalie Reesink-Peters, )