- Email: [email protected]
15 The significance of pelvic and para-aortic lymphadenectomy in the operative treatment of ovarian cancer
15 The significance of pelvic and para-aortic lymphadenectomy in the operative treatment of ovarian cancer E. B U R G H A R D T M. L A H O U S E N H. STETTNER
In 1979 we saw a patient with Stage II] ovarian cancer who had markedly enlarged, metastatically involved pelvic lymph nodes. This was surprising since little was then known about the retroperitoneal spread of ovarian cancer. Conventional wisdom held that the lymphatic drainage of the ovaries flowed directly to the para-aortic nodes (Reiffenstuhl, 1957; Plentl and Friedman, 1971). We turned to the question of lymphatic spread and tried first to elucidate by systematic lymphadenectomy--as done in cervical cancer--the extent of pelvic node involvement (Pickel et al, 1981; Burghardt et al, 1983; 1984; 1986; 1987). The results were stunning. We found positive nodes in 94 of 152 patients (61.8%) with ovarian cancer, and in 76.5% of patients with Stage III disease. This showed that the retroperitoneal spread of ovarian cancer is just as important as its abdominal spread. Wu et al (1986) have confirmed these results. In the meantime, chemotherapy for ovarian cancer has become more effective, especially after the surgical removal of as much tumour mass as possible. Consequently, surgical efforts were intensified with the goal of maximum cytoreduction (debulking). Taking the retroperitoneal spread of ovarian cancer into account--we have had patients with more than 40 positive nodes--it seemed only logical to include lymphadenectomy in the debulking procedure. Also, chemotherapy was shown unexpectedly to have no reliable effect on lymph node metastases. This would mean that a patient with positive nodes cannot be healed without their removal. We thus considered lymphadenectomy a part of cytoreduction. The question was whether this expansion of surgical treatment could improve survival. PATIENTS AND METHODS
Between 1980 and 1987, 320 patients with ovarian cancer in Stages I to IV were referred to us. Of these patients 105 underwent total abdominal BailliOre's Clinical Obstetrics and @naecology--Vol. 3, No. 1, March 1989
157
158
E. BURGHARDT ET AL
hysterectomy, bilateral salpingo-oophorectomy and omentectomy, and as complete a resection of malignant disease as possible (debulking). Ninetyfive patients underwent debulking and pelvic lymphadenectomy, while 57 patients underwent debulking together with pelvic and para-aortic lymphadenectomy. Thus, a lymphadenectomy was carried out in almost 50% of the patients. Thirty-seven women had a staging laparotomy, with or without oophorectomy. The poor condition of the remaining 26 women precluded even this operation (Table 1). Table 2 shows how our surgical efforts in Stage III disease have intensified. In the four years prior to 1980, 61.2% of Stage III patients had only a biopsy or an oophorectomy. Between 1980 and 1982 this figure sank to 24.5%, while 42.9% of the patients underwent lymphadenectomy. During the next three years a mere 7.5% of the patients underwent biopsy only-which has since been discontinued altogether--while 82% of the patients underwent a procedure which included lymphadenectomy. The number of patients who underwent only hysterectomy, salpingo-oophorectomy and omentectomy without lymphadenectomy (because of technical difficulties, or because of the condition of the patient) declined accordingly. This development was a result of increasing experience and determination. Table 3 shows more numbers reflecting the aggressiveness of surgery and the quality of treatment. They show how often we were able completely to clear the abdomen of gross disease and how often the tumour residuals were less than 2 cm in diameter. The table also shows how often the bowel was resected. Table 1. Ovarian cancer (Stages I-IV): programme of treatment (Graz, 1980-1987). Number of cases
Treatment Hysterectomy, BSO* Omentectomy, Debulking Pelvic lymphadenectomy Pelvic and para-aortic lymphadenectomy Biopsy (BSO) No surgery Total number of cases
(%)
105 95
32.8 29.7
57 37 26
17.8 11.6 8.1
320
100.0
47.5
*BSO = bilateral salpingo-oophorectomy, Table 2. Ovarian cancer (Stage III): changes in operative treatment
(Graz, 1976-1987). TAH+ Number BSO+ OM* of cases (Debulking) 1976-1979 1980--1982 1983-1985 1986-1987
49 49 53 46
19 (38.8%) 16 (32.6%) 10 (18.9%) 8 (17.4%)
Lymphadenectomy
Biopsy (SO) only*
21 (42.9%) 39 (73.6%) 38 (82.6%)
30 (61.2%) 12 (24.5%) 4 (7.5%) -
*TAH = total abdominal hysterectomy; BSO = bilateral salpingooophorectomy; OM = omentectomy; SO = sa!pingo-oophorectomy.
159
PELVIC AND PARA-AORTIC LYMPHADENECTOMY Table 3. Ovarian cancer (Stage III): frequency of bowel resection and size of residual tumonr (Graz, 1980-1987). Number No residual Residual tumour Residual tumour of cases tumour < 2 cm > 2 cm 98
33 (33.7%)
25 (25.5%)
4O (40.8%)
Bowel resection 32 (32.7%)
CLASSIFICATION According to the FIGO rules, ovarian cancers with positive retroperitoneal nodes belong to Stage III. This rule is debatable not only because it is an exception in the classification of genital cancers: it is also not practicable because systematic lymphadenectomy is done only at a few centres and is often limited--thus producing undependable results (Table 4). We have Table 4. Ovarian cancer (Stages I-IV): pelvic and para-aortic node involvement, and sampling vs systematic lymphadenectomy. Pelvic nodes
Aortic nodes
Averette et al (1983) Chen & Lee (1983)
24.4% 14.8%
29.3% } 37.7% _ Sampling
Own results (1987)
57.9%
50.8% Systematic lymphadenectomy
continued to classify our patients only on the basis of the intraperitoneal findings. This lets us point out the degree of intraperitoneal spread at which positive nodes must be expected. STATISTICS Survival rates and probabilities were calculated by the actuarial life table method. Recurrence-free rates were calculated with the reduced-sample method. Differences in probabilities were assessed by the log-rank test, while differences in rates and proportions were assessed by the Fisher-Yates test, and in some cases by the chi-square test. The confidence intervals were calculated with exact methods to determine the parameter of the binomial distribution. RESULTS OF OPERATIVE TREATMENT OF OVARIAN CANCER WITH LYMPHADENECTOMY
Cytoreduction, including lymphadenectomy, has improved the fate of patients at all stages of ovarian cancer (Figure 1). In Stage III, with a representative number of treated patients, 5-year survival exceeded 40%. If
160
E. BURGHARDT ET AL
80,4A
o~
v
60'
40' I3-
20 0
0
l
I
I
I
2
3
4
5
Survival (years) Figure 1. Actuarial survival after operative treatment of ovarian cancer (Stages I to IV) with lymphadenectomy (own results, 1980-1987). • Stage I (n = 28); O Stage II (n = 13); • Stage III (n=98); [-]Stage IV (n = 13).
8o.
100-
-.~
~
80- l 20-
/~//~
0
J
I
I
I
Stage IV
Stage III
Stage 1 - 11
Figure 2. The operative treatment of ovarian cancer (Stages I-IV), including lymphadenectomy: actuarial 5-year survival according to nodal status (own results, 1980-1987). [ ] lymph nodes negative; • lymph nodes positive. 100( 80
....
•
•
0
0
60
'~
40 20 0
I
0
I
!
I
2 3 Survival (years)
I
I
4
5
Figure 3. The operative treatment of ovarian cancer (Stage III), including lymphadenectomy: actuarial 5-year survival according to nodal status in the absence of residual tumour (own results, 1980-1987), • lymph nodes negative (n= 18); C) lymph nodes positive (n= 15); P=0.01.
161
PELVIC AND PARA-AORTIC LYMPHADENECTOMY
we consider Stages I and II together, and compare them with Stage III, we see that the actuarial survival rates depend on whether the nodes are positive or not (Figure 2). The same is true for Stage IV, where all patients with positive nodes died. An important determinant of survival in ovarian cancer is whether any-and how much--residual tumour is left in the abdomen at surgery. Five-year survival was high if, in addition to lymphadenectomy, the abdomen had been completely cleared of gross disease. But survival differed according to whether the nodes were positive or not (Figure 3). The same was true for patients with residual disease at surgery (Figure 4). 100~
80 ¸
60 c ¢o
~
12.
40 20 0
|
....
i
o Survival (years)
Figure 4. The operative treatment of ovarian cancer (Stage III), including lymphadenectomy: actuarial 5-year survival according to nodal status in the presence of residual tumour (own results, 1980-1987). O lymph nodes negative (n = 11); C) lymph nodes positive (n=54); P = 0.001.
THREE-YEAR SURVIVAL WITH NO EVIDENCE OF DISEASE
There is a widespread opinion that, with all the new measures in the treatment of ovarian cancer, we will be able to prolong survival but not heal the patient. Considering survival without clinical recurrence and without evidence of recurrence, by computed tomography or tumour-marker proTotal n = 49
Positive lymph nodes (n = 37)
Negative lymph nodes (n = 12)
10090" A
80"
7060-
g
50-
'= 40~- 3020" 10O.
I
I
Figure 5. The operative treatment of ovarian cancer (Stage III), including pelvic lymphadenectomy: 3-year survival with no evidence of disease (own results, 1980-1987). [ ] No evidence of disease; ~ survival with recurrence; {~ died of disease; [ ] died of other causes.
162
E. B U R G H A R D T ET A L
files, the results were remarkable. Figure 5 shows that, in Stage III, after surgery which included only pelvic lymphadenectomy, the 3-year survival rate without recurrence was 33.3 % - - e v e n though a number of patients died of other causes. Patients with negative nodes did better than those with positive nodes. Considering the outcome in the light of residual turnout at surgery, we see that the (admittedly few) patients with no residual disease and with negative nodes had a recurrence-free survival rate of about 70%. However, patients with positive nodes also had a relatively good fate (Figure 6). The results were poorer in patients with residual tumour at surgery, but even these patients had recurrence-free survival that was paradoxically longer in patients with positive nodes than in those with negative nodes (Figure 7). Total n = 16
Positive lymph nodes (n = 9)
Negative lymph nodes (n = 7)
10080"
== %
6040-
20O-
t
Figure 6. The operative treatment of ovarian cancer (Stage III), including pelviclymphadenectomy: 3-year survival with no evidence of disease in patients having no residual turnout at surgery (own results, 1980-1987). [] No evidence of disease; [ ] survival with recurrence; [] died of disease; • died of other causes. Total n = 33
Positive lymph nodes (n = 28)
Negative lymph nodes (n = 5)
10090-
80" 7080. ~= 50. ' ; 40, 30. 20. 10O-
Figure 7. The operative treatment of ovarian cancer (Stage III), including pelvic lymphadenectomy: 3-year survival with no evidence of disease in patients having residual tumour at surgery (own results, 1980--1987),[] No evidence of disease; [] survival with recurrence; [] died of disease; • died of other causes.
T H E I M P O R T A N C E OF LYMPHADENECTOMY It is too early to answer the difficult question of the role of lymphadenectomy in the surgical management of ovarian cancer and to say how much it
PELVIC AND PARA-AORTIC LYMPHADENECTOMY
163
can improve the outcome. We began the present series as a pilot study and expanded it while also intensifying our surgical efforts (Table 2). Systematic para-aortic lymphadenectomy has been done only in the last two and a half years, so our collection is not homogeneous. Randomization of the patients was out of the question as long as the pattern of retroperitoneal spread of the disease was unclear. Also, the surprisingly high incidence of lymphatic metastasis then made it unthinkable to leave positive nodes in situ for a randomized study, especially as chemotherapy had been shown to be ineffective against node metastases (Burghardt et al, 1986; 1987). As to the importance of node involvement, we found that the involvement of only a single node entailed a much better prognosis than did the involvement of a number of nodes (Figure 8). We found no correlation between the size of the node metastases and survival. Another fact indicates the value of lymphadenectomy. Of the 33 women with Stage III disease who underwent only a pelvic lymphadenectomy, and who had no gross residual tumour at surgery, ten (30.3%) had a recurrence. 100( 80. A
57.8%
60. E ¢D
40.
26.8%
r~
20.
o"
;
....
Survival (years)
Figure 8. Ovarian cancer (Stages I-IV): actuarial survival according to the number of positive nodes (own results, I980-1987). O One node involved; C) more than one node involved.
IO0~ o~
80 ' ~ 60 '
~
2O 0
0 Survival (years)
Figure 9. Ovarian cancer (Stage III), effect of lymphadenectomy (own results, 1976-1987) on actuarial survival. All cases, irrespective of treatment: (~) 1980-1983 (n = 69); • (n = 79). No lymphadenectomy: O 1976-1979 (n = 49).
1984-1987
164
E. BURGHARDT ET AL
Five of these recurrences involved the para-aortic region, and three of these five recurrences involved only the para-aortic region. However, it is not yet statistically possible to compare the results of a pelvic lymphadenectomy with those of a pelvic and para-aortic lymphadenectomy, or to evaluate those patients who had only positive para-aortic nodes. Figure 9 shows the results of all Stage III patients during the three periods under consideration, regardless of treatment. Survival improved sharply after the introduction of lymphadenectomy to a degree that cannot be explained only by the advent of cisplatin. We again emphasize that to leave in situ nodes that will be positive in a large majority of patients is incompatible with the aim of successful treatment for ovarian cancer. The mere fact that chemotherapy is ineffective against lymph node metastases forces the conclusion that therapeutic lymphadenectomy must be an integral component of the operative treatment of ovarian cancer.
SUMMARY
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 !II 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.
REFERENCES Averette HE, Lovecchio JL, Townsend PA, Sevin BU & Girtanner RE (1983) Retroperitoneal lymphatic involvement by ovarian carcinoma. In Bender HG & Beck L (eds) Carcinomaof the Ovary, pp 101-110. CancerCampaign vol. 7. Burghardt E, Pickel H, Holzer E & Lahousen M (1983) The significance of lymphadenectomy in therapy of ovarian carcinoma. American Journal of Obstetrics and Gynecology 146: 111-112. Burghardt E, Pickel H & Stettner H (1984) Management of advanced ovarian cancer. European Journal of Gynaecological Oncology 3: 155-159. Burghardt E, Pickel H, Lahousen M & Stettner H (1986) Pelvic lymphadenectomy in operative treatment of ovarian cancer. American Journal of Obstetrics and Gynecology 155: 315319.
PELVIC AND PARA-AORTIC LYMPHADENECTOMY
165
Burghardt E, Lahousen M & Stettner H (1987) The role of lymphadenectomy in the treatment of ovarian cancer. In Sharp F & Soutter WP (eds) Ovarian Cancer--the Way Ahead, pp 257-267. London: The Chamelon Press Ltd. Chen SS & Lee L (1983) Incidence of para-aortic and pelvic lymph node metastases in epithelial carcinoma of the ovary. Gynecologic Oncotogy 16- 95-100. Pickel H, Lahousen M & Holzer E (1981) Zur Bedeutung der Lymphadenektomie ffir die Therapie des Ovarialkarzinoms. Geburtshilfe und Frauenheilkunde 41; 841-843. Plentl A A & Friedman EA (1971) Lymphatic system of the female genitalia. In Friedman EA (ed.) Major Problems in Obstetrics and Gynecology, p 173. Philadelphia: WB Sauuders, Reiffenstuhl G (1957) Das Lymphsystem des weiblichen Genitale. Vienna: Urban & Schwarzenberg 184pp. Wu PCh, Qu JY, Lang JH, Huang RL, Tang MY & Lian LJ (1986) Lymph node metastasis of ovarian cancer: a preliminary survey of 74 cases of lymphadenectomy. American Journal of Obstetrics and Gynecology 155; 1103-1108.
In 1979 we saw a patient with Stage II] ovarian cancer who had markedly enlarged, metastatically involved pelvic lymph nodes. This was surprising since little was then known about the retroperitoneal spread of ovarian cancer. Conventional wisdom held that the lymphatic drainage of the ovaries flowed directly to the para-aortic nodes (Reiffenstuhl, 1957; Plentl and Friedman, 1971). We turned to the question of lymphatic spread and tried first to elucidate by systematic lymphadenectomy--as done in cervical cancer--the extent of pelvic node involvement (Pickel et al, 1981; Burghardt et al, 1983; 1984; 1986; 1987). The results were stunning. We found positive nodes in 94 of 152 patients (61.8%) with ovarian cancer, and in 76.5% of patients with Stage III disease. This showed that the retroperitoneal spread of ovarian cancer is just as important as its abdominal spread. Wu et al (1986) have confirmed these results. In the meantime, chemotherapy for ovarian cancer has become more effective, especially after the surgical removal of as much tumour mass as possible. Consequently, surgical efforts were intensified with the goal of maximum cytoreduction (debulking). Taking the retroperitoneal spread of ovarian cancer into account--we have had patients with more than 40 positive nodes--it seemed only logical to include lymphadenectomy in the debulking procedure. Also, chemotherapy was shown unexpectedly to have no reliable effect on lymph node metastases. This would mean that a patient with positive nodes cannot be healed without their removal. We thus considered lymphadenectomy a part of cytoreduction. The question was whether this expansion of surgical treatment could improve survival. PATIENTS AND METHODS
Between 1980 and 1987, 320 patients with ovarian cancer in Stages I to IV were referred to us. Of these patients 105 underwent total abdominal BailliOre's Clinical Obstetrics and @naecology--Vol. 3, No. 1, March 1989
157
158
E. BURGHARDT ET AL
hysterectomy, bilateral salpingo-oophorectomy and omentectomy, and as complete a resection of malignant disease as possible (debulking). Ninetyfive patients underwent debulking and pelvic lymphadenectomy, while 57 patients underwent debulking together with pelvic and para-aortic lymphadenectomy. Thus, a lymphadenectomy was carried out in almost 50% of the patients. Thirty-seven women had a staging laparotomy, with or without oophorectomy. The poor condition of the remaining 26 women precluded even this operation (Table 1). Table 2 shows how our surgical efforts in Stage III disease have intensified. In the four years prior to 1980, 61.2% of Stage III patients had only a biopsy or an oophorectomy. Between 1980 and 1982 this figure sank to 24.5%, while 42.9% of the patients underwent lymphadenectomy. During the next three years a mere 7.5% of the patients underwent biopsy only-which has since been discontinued altogether--while 82% of the patients underwent a procedure which included lymphadenectomy. The number of patients who underwent only hysterectomy, salpingo-oophorectomy and omentectomy without lymphadenectomy (because of technical difficulties, or because of the condition of the patient) declined accordingly. This development was a result of increasing experience and determination. Table 3 shows more numbers reflecting the aggressiveness of surgery and the quality of treatment. They show how often we were able completely to clear the abdomen of gross disease and how often the tumour residuals were less than 2 cm in diameter. The table also shows how often the bowel was resected. Table 1. Ovarian cancer (Stages I-IV): programme of treatment (Graz, 1980-1987). Number of cases
Treatment Hysterectomy, BSO* Omentectomy, Debulking Pelvic lymphadenectomy Pelvic and para-aortic lymphadenectomy Biopsy (BSO) No surgery Total number of cases
(%)
105 95
32.8 29.7
57 37 26
17.8 11.6 8.1
320
100.0
47.5
*BSO = bilateral salpingo-oophorectomy, Table 2. Ovarian cancer (Stage III): changes in operative treatment
(Graz, 1976-1987). TAH+ Number BSO+ OM* of cases (Debulking) 1976-1979 1980--1982 1983-1985 1986-1987
49 49 53 46
19 (38.8%) 16 (32.6%) 10 (18.9%) 8 (17.4%)
Lymphadenectomy
Biopsy (SO) only*
21 (42.9%) 39 (73.6%) 38 (82.6%)
30 (61.2%) 12 (24.5%) 4 (7.5%) -
*TAH = total abdominal hysterectomy; BSO = bilateral salpingooophorectomy; OM = omentectomy; SO = sa!pingo-oophorectomy.
159
PELVIC AND PARA-AORTIC LYMPHADENECTOMY Table 3. Ovarian cancer (Stage III): frequency of bowel resection and size of residual tumonr (Graz, 1980-1987). Number No residual Residual tumour Residual tumour of cases tumour < 2 cm > 2 cm 98
33 (33.7%)
25 (25.5%)
4O (40.8%)
Bowel resection 32 (32.7%)
CLASSIFICATION According to the FIGO rules, ovarian cancers with positive retroperitoneal nodes belong to Stage III. This rule is debatable not only because it is an exception in the classification of genital cancers: it is also not practicable because systematic lymphadenectomy is done only at a few centres and is often limited--thus producing undependable results (Table 4). We have Table 4. Ovarian cancer (Stages I-IV): pelvic and para-aortic node involvement, and sampling vs systematic lymphadenectomy. Pelvic nodes
Aortic nodes
Averette et al (1983) Chen & Lee (1983)
24.4% 14.8%
29.3% } 37.7% _ Sampling
Own results (1987)
57.9%
50.8% Systematic lymphadenectomy
continued to classify our patients only on the basis of the intraperitoneal findings. This lets us point out the degree of intraperitoneal spread at which positive nodes must be expected. STATISTICS Survival rates and probabilities were calculated by the actuarial life table method. Recurrence-free rates were calculated with the reduced-sample method. Differences in probabilities were assessed by the log-rank test, while differences in rates and proportions were assessed by the Fisher-Yates test, and in some cases by the chi-square test. The confidence intervals were calculated with exact methods to determine the parameter of the binomial distribution. RESULTS OF OPERATIVE TREATMENT OF OVARIAN CANCER WITH LYMPHADENECTOMY
Cytoreduction, including lymphadenectomy, has improved the fate of patients at all stages of ovarian cancer (Figure 1). In Stage III, with a representative number of treated patients, 5-year survival exceeded 40%. If
160
E. BURGHARDT ET AL
80,4A
o~
v
60'
40' I3-
20 0
0
l
I
I
I
2
3
4
5
Survival (years) Figure 1. Actuarial survival after operative treatment of ovarian cancer (Stages I to IV) with lymphadenectomy (own results, 1980-1987). • Stage I (n = 28); O Stage II (n = 13); • Stage III (n=98); [-]Stage IV (n = 13).
8o.
100-
-.~
~
80- l 20-
/~//~
0
J
I
I
I
Stage IV
Stage III
Stage 1 - 11
Figure 2. The operative treatment of ovarian cancer (Stages I-IV), including lymphadenectomy: actuarial 5-year survival according to nodal status (own results, 1980-1987). [ ] lymph nodes negative; • lymph nodes positive. 100( 80
....
•
•
0
0
60
'~
40 20 0
I
0
I
!
I
2 3 Survival (years)
I
I
4
5
Figure 3. The operative treatment of ovarian cancer (Stage III), including lymphadenectomy: actuarial 5-year survival according to nodal status in the absence of residual tumour (own results, 1980-1987), • lymph nodes negative (n= 18); C) lymph nodes positive (n= 15); P=0.01.
161
PELVIC AND PARA-AORTIC LYMPHADENECTOMY
we consider Stages I and II together, and compare them with Stage III, we see that the actuarial survival rates depend on whether the nodes are positive or not (Figure 2). The same is true for Stage IV, where all patients with positive nodes died. An important determinant of survival in ovarian cancer is whether any-and how much--residual tumour is left in the abdomen at surgery. Five-year survival was high if, in addition to lymphadenectomy, the abdomen had been completely cleared of gross disease. But survival differed according to whether the nodes were positive or not (Figure 3). The same was true for patients with residual disease at surgery (Figure 4). 100~
80 ¸
60 c ¢o
~
12.
40 20 0
|
....
i
o Survival (years)
Figure 4. The operative treatment of ovarian cancer (Stage III), including lymphadenectomy: actuarial 5-year survival according to nodal status in the presence of residual tumour (own results, 1980-1987). O lymph nodes negative (n = 11); C) lymph nodes positive (n=54); P = 0.001.
THREE-YEAR SURVIVAL WITH NO EVIDENCE OF DISEASE
There is a widespread opinion that, with all the new measures in the treatment of ovarian cancer, we will be able to prolong survival but not heal the patient. Considering survival without clinical recurrence and without evidence of recurrence, by computed tomography or tumour-marker proTotal n = 49
Positive lymph nodes (n = 37)
Negative lymph nodes (n = 12)
10090" A
80"
7060-
g
50-
'= 40~- 3020" 10O.
I
I
Figure 5. The operative treatment of ovarian cancer (Stage III), including pelvic lymphadenectomy: 3-year survival with no evidence of disease (own results, 1980-1987). [ ] No evidence of disease; ~ survival with recurrence; {~ died of disease; [ ] died of other causes.
162
E. B U R G H A R D T ET A L
files, the results were remarkable. Figure 5 shows that, in Stage III, after surgery which included only pelvic lymphadenectomy, the 3-year survival rate without recurrence was 33.3 % - - e v e n though a number of patients died of other causes. Patients with negative nodes did better than those with positive nodes. Considering the outcome in the light of residual turnout at surgery, we see that the (admittedly few) patients with no residual disease and with negative nodes had a recurrence-free survival rate of about 70%. However, patients with positive nodes also had a relatively good fate (Figure 6). The results were poorer in patients with residual tumour at surgery, but even these patients had recurrence-free survival that was paradoxically longer in patients with positive nodes than in those with negative nodes (Figure 7). Total n = 16
Positive lymph nodes (n = 9)
Negative lymph nodes (n = 7)
10080"
== %
6040-
20O-
t
Figure 6. The operative treatment of ovarian cancer (Stage III), including pelviclymphadenectomy: 3-year survival with no evidence of disease in patients having no residual turnout at surgery (own results, 1980-1987). [] No evidence of disease; [ ] survival with recurrence; [] died of disease; • died of other causes. Total n = 33
Positive lymph nodes (n = 28)
Negative lymph nodes (n = 5)
10090-
80" 7080. ~= 50. ' ; 40, 30. 20. 10O-
Figure 7. The operative treatment of ovarian cancer (Stage III), including pelvic lymphadenectomy: 3-year survival with no evidence of disease in patients having residual tumour at surgery (own results, 1980--1987),[] No evidence of disease; [] survival with recurrence; [] died of disease; • died of other causes.
T H E I M P O R T A N C E OF LYMPHADENECTOMY It is too early to answer the difficult question of the role of lymphadenectomy in the surgical management of ovarian cancer and to say how much it
PELVIC AND PARA-AORTIC LYMPHADENECTOMY
163
can improve the outcome. We began the present series as a pilot study and expanded it while also intensifying our surgical efforts (Table 2). Systematic para-aortic lymphadenectomy has been done only in the last two and a half years, so our collection is not homogeneous. Randomization of the patients was out of the question as long as the pattern of retroperitoneal spread of the disease was unclear. Also, the surprisingly high incidence of lymphatic metastasis then made it unthinkable to leave positive nodes in situ for a randomized study, especially as chemotherapy had been shown to be ineffective against node metastases (Burghardt et al, 1986; 1987). As to the importance of node involvement, we found that the involvement of only a single node entailed a much better prognosis than did the involvement of a number of nodes (Figure 8). We found no correlation between the size of the node metastases and survival. Another fact indicates the value of lymphadenectomy. Of the 33 women with Stage III disease who underwent only a pelvic lymphadenectomy, and who had no gross residual tumour at surgery, ten (30.3%) had a recurrence. 100( 80. A
57.8%
60. E ¢D
40.
26.8%
r~
20.
o"
;
....
Survival (years)
Figure 8. Ovarian cancer (Stages I-IV): actuarial survival according to the number of positive nodes (own results, I980-1987). O One node involved; C) more than one node involved.
IO0~ o~
80 ' ~ 60 '
~
2O 0
0 Survival (years)
Figure 9. Ovarian cancer (Stage III), effect of lymphadenectomy (own results, 1976-1987) on actuarial survival. All cases, irrespective of treatment: (~) 1980-1983 (n = 69); • (n = 79). No lymphadenectomy: O 1976-1979 (n = 49).
1984-1987
164
E. BURGHARDT ET AL
Five of these recurrences involved the para-aortic region, and three of these five recurrences involved only the para-aortic region. However, it is not yet statistically possible to compare the results of a pelvic lymphadenectomy with those of a pelvic and para-aortic lymphadenectomy, or to evaluate those patients who had only positive para-aortic nodes. Figure 9 shows the results of all Stage III patients during the three periods under consideration, regardless of treatment. Survival improved sharply after the introduction of lymphadenectomy to a degree that cannot be explained only by the advent of cisplatin. We again emphasize that to leave in situ nodes that will be positive in a large majority of patients is incompatible with the aim of successful treatment for ovarian cancer. The mere fact that chemotherapy is ineffective against lymph node metastases forces the conclusion that therapeutic lymphadenectomy must be an integral component of the operative treatment of ovarian cancer.
SUMMARY
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 !II 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.
REFERENCES Averette HE, Lovecchio JL, Townsend PA, Sevin BU & Girtanner RE (1983) Retroperitoneal lymphatic involvement by ovarian carcinoma. In Bender HG & Beck L (eds) Carcinomaof the Ovary, pp 101-110. CancerCampaign vol. 7. Burghardt E, Pickel H, Holzer E & Lahousen M (1983) The significance of lymphadenectomy in therapy of ovarian carcinoma. American Journal of Obstetrics and Gynecology 146: 111-112. Burghardt E, Pickel H & Stettner H (1984) Management of advanced ovarian cancer. European Journal of Gynaecological Oncology 3: 155-159. Burghardt E, Pickel H, Lahousen M & Stettner H (1986) Pelvic lymphadenectomy in operative treatment of ovarian cancer. American Journal of Obstetrics and Gynecology 155: 315319.
PELVIC AND PARA-AORTIC LYMPHADENECTOMY
165
Burghardt E, Lahousen M & Stettner H (1987) The role of lymphadenectomy in the treatment of ovarian cancer. In Sharp F & Soutter WP (eds) Ovarian Cancer--the Way Ahead, pp 257-267. London: The Chamelon Press Ltd. Chen SS & Lee L (1983) Incidence of para-aortic and pelvic lymph node metastases in epithelial carcinoma of the ovary. Gynecologic Oncotogy 16- 95-100. Pickel H, Lahousen M & Holzer E (1981) Zur Bedeutung der Lymphadenektomie ffir die Therapie des Ovarialkarzinoms. Geburtshilfe und Frauenheilkunde 41; 841-843. Plentl A A & Friedman EA (1971) Lymphatic system of the female genitalia. In Friedman EA (ed.) Major Problems in Obstetrics and Gynecology, p 173. Philadelphia: WB Sauuders, Reiffenstuhl G (1957) Das Lymphsystem des weiblichen Genitale. Vienna: Urban & Schwarzenberg 184pp. Wu PCh, Qu JY, Lang JH, Huang RL, Tang MY & Lian LJ (1986) Lymph node metastasis of ovarian cancer: a preliminary survey of 74 cases of lymphadenectomy. American Journal of Obstetrics and Gynecology 155; 1103-1108.