Authors' response to: Lymphadenectomy in ovarian cancer–an overrated procedure?

Author's Reply Sir, It is true that there are few prospective, randomised, well-performed controlled studies evaluating the role of lymphadenectomy in all gynaecological cancers and, in particular, ovarian cancer. Institutional and individual bias has led us to this point and this letter by Dietl et al.1 is an example of such bias. Phrases such as ‘an overrated procedure’, together with summarising the only randomised prospective trial as failing to show a positive effect of systematic lymphadenectomy (SL) on survival, are biased and misleading. Panici and colleagues compared SL with the removal of bulky lymph nodes, and demonstrated a significant difference in progression-free interval (PFI) in the SL group, a measurement many call the most significant in the evaluation of treatment strategies in ovarian cancer.2 The Women's Cancer Center reached the same conclusion regarding the importance of removing macroscopically involved lymph nodes.3 Others have described a similar difficulty in differentiating positive from negative nodes. du Bois and colleagues, in a combined exploratory analysis of all three prospective randomised trials concerning this topic, pointed out the shortcomings of the report by Panici and colleagues, the most glaring of which was the small number of patients with complete resection of all intraperitoneal disease. Furthermore, du Bois and colleagues3 and Ditto and colleagues4 emphasised the importance of SL in ‘early-stage’ disease. Dietl and colleagues do not seem to acknowledge that the same surgical exploration is required whether or not the nodes are removed. Dietl and collegues then mistakenly attribute the difference in survival reported by Burghardt and colleagues to a change in chemotherapy and the introduction of cisplatin-based regimens. This is not supported by an in-depth analysis presented at the Toronto Gynecologic Oncology Group (GOG) Educational Symposium (July 1998). In analysing patients with no visible disease after surgery in GOG studies #25 (melphalan ± Corynebacterium parvum), #52 [cisplatin–adriamycin–cyclophosphamide (PAC) versus cisplatin/cytoxan (PC)] and #114 [intraperitoneal versus intravenous cisplatin/taxol (PT)], the median survivals were 53 months, 55 and 62 months, and 52 months in the intravenous arm of #114, respectively. Dietl and colleagues have failed to understand that, at the same time as platinum-based regimens were introduced, the definition of the optimal residual changed from 3 to 1 cm. Bias, not science, has resulted in ascribing the benefits in overall survival to one (chemotherapy) of the two variables that have changed simultaneously. So much for another ‘undisputed truth’. Dietl and colleagues also state, ‘For ethical reasons, the considerable side-effects of systematic LNE, such as greatly enhanced blood loss, prolonged duration of surgery and a more complicated healing process with additional risks for pulmonary embolism and ileus, must be weighed against the benefits associated with the procedure.’. Panici and colleagues make no mention of an increased risk of pulmonary embolism, but, rather, state that there was no difference in perioperative complications. Moreover, one must question the clinical significance of transfusing 71% of patients versus 60% despite a ‘significant P value’. More unsubstantiated claims include, ‘Patients with ovarian cancer, however, rarely die from lymphogenic metastasis. Instead, the intra-abdominal disease represents the major problem’, and concerns regarding the detriment of removing ‘immunologically relevant’ tissue. In our experience, as we become more successful in achieving complete cytoreduction, more patients are dying from distant disease, and the exact immunological relevance of this group of lymph nodes clearly requires more definition. I would strongly suggest that the authors making such claims rethink their questioning of retrospective data and direct that circumspection to completely unsupported statements such as these. Perhaps we should focus on the many unanswered questions regarding the molecular nature of lymph node metastases. Berek and colleagues5 reported over 25 years ago that approximately 50% of patients with negative intraperitoneal findings, on second examination, have positive lymph nodes; Kimball and colleagues6 demonstrated that a higher than suspected percentage of lymph node metastases are diploid, not aneuploid; and McAlpine and colleagues7 and others have further demonstrated the heterogeneous nature of metastatic ovarian cancer. We should be congratulating, not criticising, Burghardt and colleagues for bringing these issues to the forefront.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.