Definition of compartment-based radical surgery in uterine cancer: modified radical hysterectomy in intermediate/high-risk endometrial cancer using peritoneal mesometrial resection (PMMR) by M Höckel translated to robotic surgery

Treatment of endometrial cancer is primarily surgical. However, there is ongoing debate on the extent of surgical radicalness, predominantly with regard to lymphadenectomy (LNE). In addition, adjuvant radiotherapy has been shown to improve local control but does not increase survival probability. Indeed, adjuvant chemotherapy seems to increase survival probability in patients with high-risk endometrial cancer[1, 2]. Surgical treatment of endometrial cancer can be performed with lower risk but apparently unaltered oncological efficiency by minimally invasive surgery. This also holds true for minimally invasive approaches and robotic surgery[3‐6].

Although there are many reports of the results of different surgical approaches, a standardization of surgical treatment for intermediate/high-risk endometrial cancer is missing. Standardization is crucial for the evaluation of locoregional tumor control by surgery alone, as it has been proposed for cervical cancer[7‐9]. Applying the findings of basic research on boundary formation and maintenance in tissue development[10], as well as the results of clinical research on ontogenetic compartments of cancer spread in endometrial cancer[11], the compartments of local and regional cancer spread can be defined and surgical techniques adapted to remove the compartments at risk entirely. In analogy to cervical cancer, this may also impact upon locoregional cancer recurrence in endometrial cancer without any additional adjuvant radiotherapy. As proposed by M Höckel, the technique of TMMR can be modified and adapted to localization and lymph drainage of endometrial cancer, with regard to the paramesonephric-mesonephric-Müllerian tubercle complex-derived tissue compartments (personal communication).

In this paper, we describe the resulting technique of peritoneal mesometrial resection (PMMR), which has been translated from open to robotic surgery (rPMMR). Clinical outcome data are reported first.

In total, 42 patients received robotically-assisted laparoscopic surgery. Thirty-nine patients received rPMMR with bilateral salpingo-oophorectomy (n = 12), additional pelvic tLNE (n = 8) or pelvic and periaortic LNE (n = 19). Three patients underwent completion of surgery following hysterectomy and diagnosis of endometrial cancer including resection of mesometria. Infundibulopelvic ligaments and tLNE were categorized as rPMMR with tLNE. Another two patients received immediate secondary robotic surgery for periaortic LNE in case of clinically inconspicuous but histologically verified positive pelvic nodes, and were also categorized as rPMMR with pelvic and periaortic tLNE. In six cases, resection of ligamentous mesometria was added due to suspicion of additional cervical involvement. The remaining three patients received robotic total laparoscopic hysterectomy (rTLH) only for stage IA endometrial cancer.

All operations were completed endoscopically without switching to open surgery. No intraoperative complications occurred. All tumors were histologically resected R0, but one tumor was diagnosed Rx (pT3b, pN1 (24/46)). Mean lymph node count was 33 (± 11) and 35 (± 10) for pelvic and pelvic/periaortic LNE, respectively. Overall, the mean decrease (± SD) of perioperative hemoglobin concentration (on the first postoperative day) was 2.4 g/dL (± 1.3 g/dL), with the lowest difference in rTLH (0.7 g/dL (± 0.6 g/dL)). The decrease following rPMMR was 2.1 g/dL (± 1.0 g/dL), with pelvic tLNE 2.6 g/dL ± 1.1 g/dL; essentially the same for pelvic and periaortic tLNE accounting for 2.6 g/dL ± 1.5 g/dL. Only one patient required transfusion following rPMMR and pelvic tLNE, with a preoperative hemoglobin concentration of 9.0 g/dL (2.4%).

Postoperative complications were relatively rare and corresponded to a total complication rate of 12%. One patient had superficial vein thrombosis, another patient had deep vein thrombosis, one patient developed an infected lymph cyst, another patient developed transient postoperative aphasia (all following rPMMR with pelvic and periaortic tLNE), and one patient had transient dysfunction of micturition following rPMMR only.

Endometrial cancer patients with stage III tumors, serous histology and/or positive lymph nodes were treated with six cycles of adjuvant platinum-based chemotherapy (n = 11). No patients received adjuvant external beam radiotherapy.

Two patients with endometrial cancer developed recurrent disease. The first patient with endometrial cancer recurrence was a 26-year-old woman with obesity (BMI 57 kg/m²). This was the only patient who required transfusion, showing a preoperative hemoglobin concentration of 9.0 g/dL. The initial tumor stage was pT1a, N0 (0/29), G3 and the histopathology revealed an endometrioid tumor type. The patient did not receive adjuvant treatment. One year after initial diagnosis, the patient presented with liver metastases and there was no evidence of relapse in the pelvis. The second patient with endometrial cancer recurrence initially had a tumor stage of pT3b, N1 (24/46), G2 (endometrioid histopathology). The tumor was completely resected macroscopically, but was shown to be histologically Rx, indicating that the tumor had already transgressed compartment borders. This patient received adjuvant chemotherapy with carboplatin (area under curve (AUC) 5) and paclitaxel 175 mg/m² d1, q3w over six cycles. After 11 months, the patient presented with local pelvic recurrence and liver metastases, and was treated with palliative chemotherapy. Both patients died from metastatic disease during the observation period. There was no further recurrence or deaths during follow-up. At present, recurrence-free survival is 95% (40 out of 42 patients).

Currently, there is ongoing international debate about the extent of surgery, mode of access and correct choice of adjuvant treatment for intermediate/high-risk endometrial cancer[1‐6]. Despite prospective data showing no benefit of adjuvant irradiation therapy, with regard to overall survival for patients with endometrial cancer, no relevant change in clinical routine has been observed, at least following the first publications of the Gynecologic Oncology Group-Adjuvant Radiation for Intermediate Risk Endometrial Cancers (GOG99) and Post Operative Radiation Therapy in Endometrial Carcinoma (PPORTEC1)[16]. Thus, highly effective and potentially curative radiotherapy will be wasted in the adjuvant setting, and will not be available for treatment in cases of recurrent disease.

With regard to the recurrence pattern of endometrial cancer, approximately 30 to 50% of patients experience locoregional recurrence, while the remaining patients present with distant or concomitant distant and locoregional recurrence[17‐19]. Locoregional recurrences usually arise in the vagina as central pelvic recurrence in the Müllerian compartment or at the pelvic side wall, typically in pelvic lymph nodes. Prognosis is usually poor except for the isolated vaginal recurrence[20]. There is some evidence that more radical surgery may lower locoregional recurrence, not only in the pelvis[17, 21] but also with respect to periaortic nodes[22, 23]. However, it is difficult to compare results from clinical studies due to the large variety of surgical strategies and quality.

Thus, a standardization and clear definition of surgical procedures is urgently required, since surgery is without doubt the mainstay in the treatment of primary endometrial cancer. Data regarding LNE vary grossly between different studies. If pelvic and periaortic LNE is performed to provide regional cure by itself, then it should, at least intentionally, be complete. Additionally, the compartments at risk of lymph basins to be removed need to be exactly defined.

Regarding tLNE, we recently proposed a surgical strategy demonstrated by minimally invasive, robotically-assisted surgery[12]. On the basis of preclinical and clinical findings of ontogenetically-defined boundary formation and strict border control[10], and the clinical research on ontogenetic compartments of cancer spread in gynecological cancer[11], the local and regional compartments at risk in endometrial cancer can be defined as they have been for cervical cancer. As outlined already, there is no need for resection of the lymphatic drainage region of the lower cervical Müllerian compartment in endometrial cancer if the tumor is confined to the uterine corpus. Additionally, lymph drainage of the uterine corpus follows the ovarian vessels of the mesonephric origin directly to the periaortic nodes. Consequently, these compartments need to be cleared by tLNE in endometrial cancer.

If M Höckel’s concept of locoregional control by compartment-based surgery alone holds true for endometrial cancer, then the implementation of ontogenetic anatomy in surgical strategies is crucial to achieve best results in terms of local and regional tumor control. Following translation to robotic surgery, we demonstrate in a first feasibility analysis that this method may be reproduced and safely performed with regard to perioperative morbidity, irrespective of additional morbidity and BMI of patients. Interestingly, there was no remarkable difference in blood loss between the different approaches of rPMMR with or without LNE. However, postoperative complications were more frequent in the group with periaortic tLNE, which not surprisingly indicates that extension of surgery and operation time play a role in postoperative morbidity. There was not a single isolated locoregional recurrence despite omitting external beam radiotherapy generally.

Since the follow-up time of the patients is limited, at present we cannot conclude that PMMR and rtLNE exert excellent locoregional control without additional radiotherapy; however, up until now, the postoperative course does not seem to disprove it. Furthermore, rPMMR and rtLNE were accompanied by relatively low morbidity, with regard to the patients’ preoperative condition, BMI and surgical radicalness. Due to the excellent three-dimensional high-definition (3DHD) vision, and the high precision and control of movements, this technique can be considered outstandingly suited for complex surgery maintaining a minimally invasive approach. It is evident that at least equal radicalness may be achieved compared to open surgery. The possibility of video documentation may be beneficial concerning scientific, educational and forensic aspects.

From our experience of this first series of rPMMR with and without rtLNE in endometrial cancer, compartment-based surgery appears in an oncological manner to exert excellent locoregional control combined with a low complication rate, especially with regard to preexistent morbidity of patients.

In conclusion, we suggest that the minimally invasive approach of compartment-based oncologic surgery for endometrial cancer by robotic assistance is feasible and safe, with regard to perioperative morbidity for patients with locally-confined endometrial cancer. Herewith, we present a proposal for compartment-based, robotically-assisted surgery for intermediate/high-risk endometrial cancer with the intention of locoregional control by surgery alone. The next step will be to evaluate whether we can achieve comparable results in endometrial cancer patients to the excellent monocentric data of M Höckel in cervical cancer patients[7‐9]. Certainly, a greater number of patients, a longer follow-up time and a multicenter approach will be mandatory for sufficient analysis. This standardized and reproducible surgical protocol could be the basis for evaluation with regard to locoregional control by surgery alone, without additional radiotherapy. Reflecting that the majority of patients who will die from endometrial cancer are also diagnosed with lung and liver metastases[24], this collective of comprehensively and reproducibly staged patients may be appropriate for the most important further development of systemic treatment for distant control not compromised by concomitant radiotherapy[2]. Finally, other surgical gynecologic oncologists may refer to this technique to standardize and compare surgical results of compartment-related surgery in the treatment of intermediate/high-risk endometrial cancer.

Written informed consent was obtained from the patient for the publication of the data and accompanying images.