In surgical gynecology the concept of radicality has long been closely associated with the surgical treatment of cervical cancer. Wilhelm Alexander Freund reported on the abdominal extirpation of a carcinomatous uterus in Breslau (now Wroclaw, Poland) in 1878 [1]. His contemporaries and pupils immediately recognized that any hope of therapeutic success would require shifting the plane of resection laterally into the parametrial tissue in order to totally remove disease that had exceded the confines of the uterus [2]. Freund and his pupils also removed bulky, grossly positive lymph nodes. Emil Ries, who went from Strasbourg to the Postgraduate Medical School in Chicago, is the father of modern lymphadenectomy. In an almost prophetic presentation in Frankfurt, Germany, in March 1895, Ries laid out the theoretical basis for a systematic lymphadenectomy with removal of grossly normal node-bearing tissue [3]. He performed dissections in cadavers and dogs and postulated a therapeutic effect before actually having done such an operation in a patient. In 1897 Ries described his surgical technique in two patients [4]. He showed that grossly normal nodes could already contain cancer deposits and also pointed out that removing the lymphatic fatty tissue made it easier to remove the parametria. Notably, he dissected out the ureter all the way to the bladder, permitting him to resect a large part of the connective tissue around the uterus. In a postscript to his second publication, Ries went to some lengths to show that he gave Clarke’s chief Howard Kelly, and thus presumably Clark himself, the idea for the lymphadenectomy component of his radical operation [4]. Ries laid the groundwork for integrating lymphadenectomy with an extended hysterectomy to a truly radical operation. Later Joe Vincent Meigs [5] and Ernst Navratil built upon and carried out this same strategy. It was Ernst Wertheim’s accomplishment to systematically develop, apply, tabulate, and analyze what he called ‘‘The extended abdominal operation for carcinoma colli uteri.’’ His landmark monograph in 1911 demonstrated that it was possible to cure locally advanced cervical carcinoma with surgery [6]. He also showed that the abdominal approach was not inferior to the vaginal approach, as advocated by Friedrich Schauta, with regard to surgical complications and therapeutic results. Wertheim considered the ‘‘possibility of resecting a large extent of the parametrium and in advanced cases to isolate the severely fused organs’’ a particular advantage of the abdominal approach. Wertheim followed a systematic surgical plan and noted that dissecting out the ureters does not in itself constitute a radical operation (Fig. 1). The idea of gynecologic oncology did not yet exist when I began my gynecologic training on the recommendation of Isidor Amreich at the department in Graz chaired by Ernst Navratil in 1954. Nonetheless, cancer treatment was a major challenge at large central hospitals, demanding large material and physical resources. The central problem of gynecologic cancer surgery was still cervical cancer. Before the effects of screening and early detection methods had come into full effect, most of the patients we saw had very advanced disease. But the problem was not only the relative frequency of the disease. It was also the treatment, particularly surgery. Radiation therapy was reserved for locally or medically inoperable patients. If the carcinoma had not yet advanced to and become fixed at the pelvic wall, the goal of treatment was radical surgical excision. Even the radiation therapists, aware of their own modest results, referred patients they still considered operable to the gynecologic surgeons. Navratil was a pupil of the Schauta school, which was continued by von Peham and Amreich [7], but in Vienna he also became familiar with Wertheim’s department. Navratil’s early passion was vaginal surgery, and many noted gynecologic surgeons of the day came to Graz to see his technique of radical vaginal hysterectomy. I remember him performing his 1000th Schauta procedure in the early 1960s. I believe that Meigs, who visited our department in 1956, converted Navratil to the abdominal approach to cervical cancer. Meigs [5] had improved Wertheim’s technique by integrating it with Latzko’s anatomic approach [8] and an uncompromising pelvic lymphadenectomy. In a way, he was retracing Emil Ries’s footsteps. In 1954 surgical radicality focused on excision of the parametrial tissue and a large as possible vaginal cuff. Surgeons prided themselves on photographs of their surgical specimens with the parametrial tissue spread out and the vaginal cuff opened. In those days the histologic diagnosis of a biopsy GYNECOLOGIC ONCOLOGY 70, 172–175 (1998) ARTICLE NO. GO985069

The International Federation of Gynecology and Obstetrics (FIGO) currently defines stage IA cervical cancer as lesions invading up to 5 mm into the stroma and with no more than 7 mm width; vascular invasion does not affect the stage assignment. The Society of Gynecologic Oncology (SGO) definition of stage IA is more restrictive with regard to depth of invasion but ignores width. We reviewed 69 patients with lesions exceeding the FIGO definition of stage IA treated between 1958 and 1991; 46 patients also exceeded the SGO criteria for stage IA. The frequency of vascular invasion showed no correlation with the depth of invasion but was correlated with the width of the lesion. Treatment consisted of conization or simple hysterectomy only (n = 27), radical abdominal hysterectomy with lymphadenectomy (n = 25), radical vaginal hysterectomy (n = 13), and conization followed by radiotherapy (n = 4). No patient developed a recurrence during a follow-up of 2-35 years. Two of the 25 patients with lymphadenectomy had one positive lymph node each. The first patient had a primary lesion with 3 mm invasion and 17 mm width, no vascular invasion, and one node metastasis 2 mm in diameter; the second had a lesion with 4 mm invasion and 10 mm width, vascular invasion, and a tumor-cell embolus in the marginal sinus of a node. These results indicate that the problems involved in treating microinvasive carcinoma of the cervix also apply to cases of small stage IB disease. It will not be possible to devise a staging system that simultaneously serves as a guideline for treatment. The current FIGO classification of stage IA2 should be expanded rather than restricted.

This chapter has been extensively revised to address several important issues in the field of cervical neoplasia. These include new information about the origin of cervical squamous neoplasia and its impact on our perceptions of lesion development. This leads in to a discussion of the conundrum of lesion grading, specifically the laboratory management of lesions that fall between cervical intraepithelial neoplasia grade 1 (CIN1) and grade 3 (CIN3). Strategies for managing this problem are offered, and the concept of squamous intraepithelial lesion (SIL) of intermediate (or indeterminate) grade is unveiled. The underpinning of this concept is the absence of any biomarker that can be depended upon to segregate low-grade squamous intraepithelial lesion (LSIL) from high-grade squamous intraepithelial lesion (HSIL) in light of the widespread variations in interpretation between observers and the lack of compelling information to support p16 as either a marker of HSIL or a predictor thereof. Other important topics include new approaches to superficially invasive squamous carcinoma, new management schemes, and the promise of vaccination programs. Finally, the concept of prophylactic ablation of the squamocolumnar junction (SCJ) is addressed as another possible approach to cervical cancer risk reduction in vulnerable populations.

BACKGROUND AND METHODS:The clinical staging system of cervical cancer according to the International Federation of Gynecology and Obstetrics (FIGO) entails a large measure of subjectivity. This study analyzed the results of 1028 patients with cervical cancer at three reference centers. All patients had radical surgery, and all surgical specimens were processed as histologic giant sections with precise volumetry of the tumor.RESULTS:The interpretation of the histologic findings of parametrial invasion, vascular involvement, and lymph node involvement was found to differ somewhat among the three centers. However, all these findings were associated with tumor size. Survival rates correlated more consistently with tumor volume than with clinical or histologic stage. Five-year survival rates ranged from 91% for patients with tumors smaller than 2.5 cm3 to 70% for those with tumors 10-50 cm3. The 5-year survival rate of 24 patients with tumors larger than 50 cm3 (71% of whom had lymph nodes with positive findings) was 48%. Survival rates were identical among the three centers for patients with tumors smaller than 10 cm3, despite different degrees of surgical radicality. In contrast, more radical surgery was associated with significantly better survival rates in patients with larger tumors.CONCLUSIONS:The results of this study indicate that volumetry of the tumor permits a more accurate assessment of therapeutic results in patients with cervical cancer than does the FIGO classification. Pretherapeutic assessment of tumor volume is possible with magnetic resonance imaging. It seems that maximum parametrial resection is not necessary for patients with smaller tumors (smaller than 10 cm3), but truly radical surgery in patients with bulky tumors achieves better results than those usually expected in Stage IIb cervical cancer and at least comparable to those of radiation therapy.

One hundred eighty patients with ovarian cancer underwent complete pelvic lymphadenectomy (n = 75) or pelvic and paraaortic lymphadenectomy (n = 105). Twenty-one patients underwent a preoperative biopsy of the scalene lymph nodes. The incidence of positive lymph nodes was 24% in stage I (n = 37), 50% in stage II (n = 14), 74% in stage III (n = 114), and 73% in stage IV (n = 15). Of the 105 patients who underwent pelvic and paraaortic lymphadenectomy, 13 (12%) had positive pelvic and negative paraaortic nodes and 10 (9%) had positive paraaortic and negative pelvic nodes. Positive scalene nodes were found in four patients (19%) later shown to have stage IV disease. One hundred forty patients were studied for number of involved nodes and node groups, size of nodal metastases, residual tumor, and survival. Of the 81 patients with positive nodes, most had only one or two positive node groups or one to three positive individual nodes. A few patients had seven to eight involved node groups with up to 44 positive nodes. Greater numbers of positive nodes were found in stage III than stage IV. The size of the largest nodal metastasis was not related to the clinical stage or survival, but did correlate with the number of positive nodes. Stage III patients with no residual tumor had a significantly lower rate of lymph node involvement than those with tumor residual (P less than 0.01). Actuarial 5-year survival rates of patients with stage III disease and no, one, or more than one positive nodes were 69, 58, and 28%, respectively.

In 1985 the International Federation of Gynecology and Obstetrics (FIGO) subdivided Stage IA cervical cancer and specified metric criteria to demarcate Stage IA from Stage IB. Early stromal invasion (Stage IA1) denotes the first invasive protrusions of a carcinoma in situ into the stroma. Microcarcinomas (Stage IA2) are small cancers a number of orders of magnitude larger than Stage IA1 lesions and with a maximum depth of invasion of 5 mm and a maximum horizontal spread of 7 mm; larger lesions are classified as Stage IB. This study reviews 486 patients previously classified as having Stage IA disease. This yielded 344 Stage IA1 and 101 Stage IA2 lesions; 41 cancers were reclassified as Stage IB. Three hundred nine, 89, and 38 patients were followed for greater than or equal to 5 years. One (0.3%) patient with Stage IA1 disease re-presented with Stage IIB disease 12 years after conization. Five (5.6%) patients with Stage IA2 lesions developed invasive recurrences; three died. None of the 38 patients reclassified as having a Stage IB lesion, including 16 who were treated conservatively, developed a recurrence. The FIGO classification is not a guideline for treatment. Stage IA1 lesions can be treated conservatively, but treatment in Stage IA2 must be individualized. Risk factors such as vascular space involvement and confluency are of high sensitivity but low specificity.

The issue of surgical radicality in patients with gynecologic cancers is not seen as clearly elsewhere as it is in Austria. Systematic extirpation of the affected organ and the complete attendant lymphatic tissue can lead to excellent results. In patients with stage IIb cervical cancer, especially those with positive para-aortic nodes, surgery produces better survival rates than radiotherapy. In ovarian cancer, systemic lymphadenectomy has led to a sudden increase in survival and cure rates.

Conventional clinical staging of cervical cancer is subjective because it is based on palpatory findings and inadequate because it cannot assess the single most important prognostic factor--tumor size. To determine the exactitude of in vivo MRI measurements of tumor volume, 22 patients with invasive cervical cancer were studied before surgery. The volumes obtained by MRI correlated well (r = 0.983) with those obtained by histomorphometric analysis of the surgical specimens, but only weakly with clinical stage. MRI may provide a basis for precise classification of cervical cancer and for objective comparison of surgery and radiotherapy.

Since 1980, 95 of 320 patients with ovarian cancer of Stages I to IV have undergone pelvic lymphadenectomy. A further 57 patients underwent pelvic and para-aortic lymphadenectomy. In the last three years, 83% of patients with Stage III disease had a lymphadenectomy, and 34% had no gross residual tumour after surgery. This entailed bowel resection in 33% of cases. After cytoreduction and chemotherapy, actuarial 5-year survival in Stages I and II was 90%, and was just over 40% in Stage III. Patients with positive nodes had markedly poorer survival than did those with negative nodes. Three years after only pelvic lymphadenectomy, 36.7% of patients with Stage III disease had no clinical, radiological or biochemical evidence of disease. If the abdomen had been cleared of gross disease at surgery, 70% of the patients had no evidence of disease at 3 years. The therapeutic effect of lymphadenectomy is also reflected in the marked improvement in survival of all patients with Stage III ovarian cancer after the introduction of lymphadenectomy.

MRI can define the spread, size, and volume of clinical cervical cancers. Appropriate pulse sequences and slice thicknesses are necessary. Twenty-five patients underwent MRI tumour volumetry before radical hysterectomy. The volume obtained by MRI was compared with that obtained from the histological giant sections; the volumes agreed at a statistically significant correlation coefficient of 0.983. The volumes obtained by MRI of 13 unfixed surgical specimens correlated with their histological volumes with a statistically significant coefficient of 0.894. Tumour volumes were compared with the respective clinical stages. Clinical stage did not correlate with tumour volume. Three very large tumours were in clinical Stage Ib. Tumour size is a major prognostic factor, can be measured easily, and, as the basis for classification, is superior to FIGO staging. MRI can measure tumour volume before treatment.

Surgical staging of cervical cancer samples the retroperitoneal lymph nodes and, at some centres, the parametria. While better than subjective clinical staging, its value is limited because the results of a sampling procedure differ widely from those of a systematic lymphadenectomy. Additionally, considering the pathology of parametrial involvement, it seems unlikely that biopsy can find the majority of parametrial cancer deposits. The most precise data on the spread of cervical cancer are produced by radical hysterectomy and systematic lymhadenectomy. The tumour size has proven to be the most important prognostic criterion and therefore the best suited for patient classification. Tumour size can be measured by a number of methods. Between 1971 and 1987, 583 of 867 patients with Stage Ib to IIb cervical cancer underwent surgical treatment. Lymphadenectomy was systematic and hysterectomy included the resection of the entire parametria at the pelvic wall. In a total of 359 serial giant sections were of sufficient quality for evaluation; most were Stage IIb cases. The frequency of positive pelvic lymph nodes was 30.3% among 132 Stage Ib cases and 44.7% among Stage IIb cases. Most tumours occupied over 40% of the cervical volume. Five-year survival by clinical stage failed to show a statistically significant difference between Stages Ib and IIb. Objective classification by tumour size showed the patients with the smallest tumours to have a five-year survival rate of 92.1%. The patients with the largest still-operable tumours occupying 80% to 100% or more of the cervix still had a five-year survival rate of 65%.(ABSTRACT TRUNCATED AT 250 WORDS)

The first sharp improvement in the operative treatment of cervical cancer was the shifting of the plane of resection away from the tumour into the parametria. This permitted resection of the primary cancer with a margin of healthy tissue. Systematic studies of excised parametrial tissue, carried out around the turn of the century, showed four types of parametrial involvement: continuous, discontinuous, carcinomatosis of the parametrial lymphatics, and parametrial lymph node involvement. It is well known that histologically demonstrated parametrial involvement often contradicts the clinical stage. So-called staging laparotomies are meant to address this problem but they, too, are inadequate since most parametrial cancer deposits are microscopic and cannot be palpated. In our own studies of totally extirpated parametria, contiguous cancer spread into the parametria never exceeded 10 mm, not even in the largest still-operable tumours. Thus the theory of contiguous, direct cancer spread to the pelvic wall is wrong. Parametrial involvement usually occurred as cancer deposits in the rarely mentioned parametrial lymph nodes. Parametrial involvement correlates better with the size of the primary tumour, expressed as the tumour-cervix quotient, than with the clinical stage. The smallest tumours, without showing continuous parametrial involvement, had a 3.4% incidence of positive nodes. Thirty-five per cent of the patients with the largest tumours had positive parametrial nodes. Parametrial lymph nodes were found in 280 (78%) of 359 surgical specimens processed as giant sections. Sixty-three patients (22.5%) had positive parametrial nodes. The nodes at the pelvic wall were involved in 80% of the patients with positive parametrial nodes. The five-year survival rate was 84% if the parametria were free of disease, but it dropped to 53% with any type of parametrial involvement. Survival rates did not differ much if only the parametrial nodes or only the pelvic nodes were positive (56% and 66%, respectively). However, if both groups were positive survival dropped to 43.1%. Positive parametrial nodes can be located anywhere in the parametrium; therefore, surgery must remove the entire structure. It remains to be seen whether an exception can be made for small Stage Ib tumours, or if lymphadenectomy can be omitted in these patients. If so, radical vaginal surgery may be the treatment of choice.

Between 1971 and 1985, a total of 325 cases of cervical cancer, Stages IB to IIB, in which operation was performed were evaluated with a view toward prognostic factors and survival rates. In radical abdominal operations, a complete resection of parametrial tissue was the goal. Extensive lymphadenectomy of the pelvis was performed. Operative specimens were processed by giant sections comprising cervix, lateral parametria, and vaginal cuff. Lymph nodes were cut by step-serial sections. Exact measurements of tumor sizes were done along with investigations of parametrium and lymph nodes. Tumors were classified according to a ratio of tumor size to size of cervix. Incidence of lymph node involvement increased with tumor size, reaching a maximum of 68.3% in the group with a ratio from 70% to 80%. Direct spread into the parametrium was rarely found, even in larger tumors occupying the entire cervix. parametrial lymph nodes were most often involved; these were scattered over the entire ligament. Five-year survival rates reached 88.1% in patients with no nodal involvement and 60.9% with nodal involvement. In the latter, the results depended on the number of nodal groups involved and the diameter of metastases. Parametrial involvement alone had no influence on healing rates, but when pelvic nodes were simultaneously involved, the results were less satisfactory. Survival rates based on tumor size differed only between the group with a ratio up to 20% and the large-tumor groups, with rates ranging from 97.5% to 70.9%. There was no statistical difference between Stages IB (31.1% positive nodes) and IIB (44.1% positive nodes) with regard to survival rates (82.2% and 76.9%, respectively).

From the end of 1979 to September, 1985, radical pelvic lymphadenectomy was performed at the Graz Clinic in 123 cases of Stages IA to IV ovarian cancer following maximum debulking procedure. In 97 patients lymphadenectomy was done primarily. In 26 it was performed during a follow-up operation to chemotherapy. The frequency of pelvic node involvement was 61.8% in the total material and 78.0% in 82 cases of Stage III disease only; 75.0% positive nodes were found in Stage III after chemotherapy. Aortic nodes were positive in 41.4%, but only when pelvic nodes were also positive. The 5-year actuarial survival rate for Stage III disease was 53.0% after pelvic lymphadenectomy compared with 13.0% without. In cases with negative nodes the survival rate was 74.7%; with positive nodes the survival rate was 45.9%.