'En bloc' peritoneal mesometrial resection (PMMR) and pelvic targeted compartmental lymphadenectomy (TCL) for management of patients with endometrial cancer - feasibility and safety study of a 'new kid on the block' in robotic surgery

We retrospectively analyzed our prospectively conducted database (IRB acknowledgment No KB-49/25 from the Medical Chamber in Gdansk, Poland) of consecutive patients with EC (all molecular types) and assigned to robotic surgical treatment using the DaVinci system. Patients underwent surgery between March 30, 2023 (the first patient on the system) and November 8, 2024. A surgical team was newly trained and certified to perform robotic surgery on the DaVinci system (no previous experience in any other procedures on any robotic systems). The main surgeon is also certified, performs regularly, and takes part in the research on Total MesoMetrial Resection (TMMR) and therapeutic lymphadenectomy (tLND) ( NCT01819077 ) [ 17 ], the procedure that together with PMMR + TCL belongs to the Cancer Field Surgery in gynecologic oncology [ 18 ].We attempted to perform the PMMR + TCL on every consecutive patient with EC. If, for any reason (intraoperative findings, anatomy, anesthesia, and comorbidity issues), performance of the PMMR + TCL was deemed not possible or would last too long, we did SLND with afferent lymph vessels, or SLND only followed by robotic simple hysterectomy (rSH). We excluded patients operated for any other diagnosis than EC, or those who underwent laparoscopic or open approach surgery. The PMMR + TCL was performed as defined in previously published references [ 9 , 11 ]. Technical surgical details are described in Supplementary Material 1. The detection of lymphatic pathways marked with ICG was first observed transperitoneally. Secondly, retroperitoneally, at all steps, intact lymph vessels between the uterus and the pelvic lymph nodes marked with ICG were visualized. These steps, and all the PMMR + TCL procedure, are presented at https://youtu.be/HOjDcohtf3c . The database and medical records were searched for the type of surgery (simple hysterectomy and SLND +/- afferent lymph vessels resection, or PMMR + TCL), intra- and postoperative complications, and other medical conditions. Patients’ co-morbidities and age were calculated together using a modified (malignancy excluded) Charlson Index [ 19 ]. Previous abdominal surgeries were considered in the analysis of feasibility and complication rate. Patient was classified as having a previous surgery if she had undergone at least one abdominal surgery for any reason, except cesarean section. For statistical calculations, the modified Charlson index values were presented as three ranges: 0–1, 2–3, and ≥4 [ 19 ]. Postoperative complications (within 30 days after the surgery) were presented descriptively and graded according to the Clavien-Dindo classification [ 20 ]. For staging, we used the TNM classification [ 21 ]. Histological and molecular analyses were performed according to guidelines [ 22 – 25 ] with ultrastaging of sentinel (targeted) lymph nodes. No modification of the surgical procedure and its extent was done based on molecular disease subtype. Oncological management after surgery was established on a multidisciplinary team board according to international guidelines [ 2 , 3 ]. Although we aimed and performed the PMMR + TCL, we did not follow the cancer field surgery concept for decisions about the adjuvant treatment, because both the surgical procedure and the robotic approach were new to us, and we did not want to bias the conventional management. The relationship between the type of surgical procedure and selected pathological-clinical parameters was evaluated using Pearson’s Chi2 and Kruskal-Wallis ANOVA tests. The significance level for the calculation was 0.05. Statistical analysis was performed using the Statistica software package, version 13 (TIBCO Software Inc., 2017). http://statistica.io ).

Patients’ median age was 65 years, median modified Charlson index and BMI, 2 and 31.2 m 2 /kg, respectively. EC radiological stage was T1/2 N0 M0. More details on patients and disease are presented in Table  1 . Table 1 Patients’ characteristics Variable Value [median (range)] [ n (%)] Age 65 (31–88) Modified Charlson index 2 (0–7) BMI [m 2 /kg] 31.2 (20.5–45.7) Radiological T stage: 1a 81 (68.6%) 1b 24 (20.3%) 2 13 (11.0%) Radiological N stage: 0 118 (100%) Pathological T stage (pT): 1a 1b 2 74 (62.7%) 18 (15.3%) 26 (22.0%) Pathological N stage (pN): pN0 pN1 112 (94.9%) 6 (5.1%) Histology (final): Endometrioid Serous Mixed type 114 (96.6%) 2 (1.7%) 2 (1.7%) Grade (final): G1 G2 G3 67 (56.8%) 40 (33.9%) 11 (9.3%) LVSI negative positive missing value 87 (73.7%) 27 (22.9%) 4 (3.4%) PNI negative positive missing value 111 (94.1%) 0 7 (5.9%) Molecular type POLE mutant MMRd p53 mutated NSMP missing value 5 (4.2%) 21 (17.8%) 20 (17.0%) 64 (54.2%) 8 (6.8%) Abbreviations. BMI body mass index,  LVSI lympho-vascular space invasion,  MMRd mismatch repair deficiency,  PNI perineural invasion Patients’ characteristics Pathological T stage (pT): 1a 1b 2 74 (62.7%) 18 (15.3%) 26 (22.0%) Pathological N stage (pN): pN0 pN1 112 (94.9%) 6 (5.1%) Histology (final): Endometrioid Serous Mixed type 114 (96.6%) 2 (1.7%) 2 (1.7%) Grade (final): G1 G2 G3 67 (56.8%) 40 (33.9%) 11 (9.3%) LVSI negative positive missing value 87 (73.7%) 27 (22.9%) 4 (3.4%) PNI negative positive missing value 111 (94.1%) 0 7 (5.9%) Molecular type POLE mutant MMRd p53 mutated NSMP missing value 5 (4.2%) 21 (17.8%) 20 (17.0%) 64 (54.2%) 8 (6.8%) Abbreviations. BMI body mass index,  LVSI lympho-vascular space invasion,  MMRd mismatch repair deficiency,  PNI perineural invasion Surgical procedures’ details are presented in Tables  2 and 3 . All but four surgeries were performed using three instruments (monopolar curved scissors, Maryland bipolar forceps, and ProGrasp forceps). The needle driver (the fourth instrument) was used in four patients. Table 2 Surgical procedure characteristics (all patients, n  = 118) Variable Value [median (range)] [ n (%)] Operative time (skin-to-skin) (min.) 213 (127–352) Number of instruments used 3 (3–4) Duration of postoperative hospital stay (days) 1 (1–3)* Type of surgery PMMR + TCL rSH + SLND with afferent lymphatic vessels rSH + SLND alone Other (RH with SLND) Other (RH with systematic pelvic lymphadenectomy) 79 (66.9%) 17 (14.4%) 20 (16.9%) 1 (0.8%) 1 (0.8%) Type of lymph node dissection TCL SLND with afferent lymphatic vessels SLND only Systematic pelvic (single side, non-display with ICG) 79 (66.9%) 17 (14.4%) 20 (16.9%) 6 (5.1%) Number of pelvic lymph nodes resected (both sides) 4 (2–26) Abbreviations: PMMR Peritoneal MesoMetrial resection, RH radical hysterectomy (type C1), SLND sentinel lymph nodes dissection, TCL targeted compartmental lymphadenectomy, rSH robotic simplehysterectomy Footnote: *, one patient with conversion to laparotomy excluded from this calculation (10 days) Surgical procedure characteristics (all patients, n  = 118) Type of surgery PMMR + TCL rSH + SLND with afferent lymphatic vessels rSH + SLND alone Other (RH with SLND) Other (RH with systematic pelvic lymphadenectomy) 79 (66.9%) 17 (14.4%) 20 (16.9%) 1 (0.8%) 1 (0.8%) Type of lymph node dissection TCL SLND with afferent lymphatic vessels SLND only Systematic pelvic (single side, non-display with ICG) 79 (66.9%) 17 (14.4%) 20 (16.9%) 6 (5.1%) Abbreviations: PMMR Peritoneal MesoMetrial resection, RH radical hysterectomy (type C1), SLND sentinel lymph nodes dissection, TCL targeted compartmental lymphadenectomy, rSH robotic simplehysterectomy Footnote: *, one patient with conversion to laparotomy excluded from this calculation (10 days) Table 3 Clinical and pathological characteristics stratified into types of surgery ( n  = 116) Procedure Variable PMMR + TCL ( n  = 79) rSH + SLND with afferent lymphatic vessels ( n  = 17) rSH + SLND only ( n  = 20) p BMI [m 2 /kg, median (range)] 29.9 (20.5–40.5) 31.6 (25.6–45.7) 38.1 (21.8–45.7) 0.03 Age [years, median (range)] 62 (31–78) 73 (65–86) 66.5 (36–88) < 0.001* Modified Charlson index, n (%) 0–1 2–3 ≥4 31 (39.0%) 41 (52.0%) 7 (9.0%) 3 (15.0%) 8 (40.0%) 9 (45.0%) 0 (0) 8 (47.0%) 9 (53.0%) < 0.001 † Previous abdominal surgeries § 23 (29.1%) 5 (29.4%) 11 (55.0%) NS † Skin-to skin time [minutes, median (range)] 222.5 (150–328) 205 (135–265) 192.5 (127–352) 0.03* Number of right pelvic LNs removed [median (range)] 4 (1–12) 3 (1–10) 3 (0–7) 0.009* Number of left pelvic LNs removed [median (range)] 4 (1–9) 3(1–8) 3 (1–6) NS* pN1 n (%) Yes No 4 (5.0%) 75 (95.0%) 1 (5.0%) 19 (95.0%) 1 (6.0%) 16 (94.0%) NS † Footnote: *, ANOVA Kruskal-Wallis test; †, Chi 2 Pearsona test; §, any type of abdominal surgery, excluding cesarean section Abbreviations: BMI body mass index, LND  lymphadenectomy, LNs  lymph nodes, NS  not statistically significant, PMMR  peritoneal mesometrial resection, pN1  positive (metastatic) lymph node (histologically), the TNM staging, RH  radical hysterectomy (type C1), SLND  sentinel lymph nodes dissection, TCL  targeted compartmental lymphadenectomy, rSH  robotic simple hysterectomy Clinical and pathological characteristics stratified into types of surgery ( n  = 116) 29.9 (20.5–40.5) 31.6 (25.6–45.7) 38.1 (21.8–45.7) 0–1 2–3 ≥4 31 (39.0%) 41 (52.0%) 7 (9.0%) 3 (15.0%) 8 (40.0%) 9 (45.0%) 0 (0) 8 (47.0%) 9 (53.0%) 222.5 (150–328) 205 (135–265) 192.5 (127–352) pN1 n (%) Yes No 4 (5.0%) 75 (95.0%) 1 (5.0%) 19 (95.0%) 1 (6.0%) 16 (94.0%) Footnote: *, ANOVA Kruskal-Wallis test; †, Chi 2 Pearsona test; §, any type of abdominal surgery, excluding cesarean section Abbreviations: BMI body mass index, LND  lymphadenectomy, LNs  lymph nodes, NS  not statistically significant, PMMR  peritoneal mesometrial resection, pN1  positive (metastatic) lymph node (histologically), the TNM staging, RH  radical hysterectomy (type C1), SLND  sentinel lymph nodes dissection, TCL  targeted compartmental lymphadenectomy, rSH  robotic simple hysterectomy The PMMR + TCL was performed on 79 patients, the rSH + SLND with afferent lymphatic vessels and the rSH + SLND alone were performed on 17 and 20 patients, respectively. Among these three procedures, the median number of removed pelvic lymph nodes (both sides) was 8, 6, and 6, respectively. Patients in whom the PMMR + TCL was performed were younger, had lower BMI and the modified Charlson Index as compared to those who underwent rSH + SLND +/- afferent lymphatic vessels (Table  3 ). Two patients underwent other procedures: a radical hysterectomy (type C1) (because of radiological cervical involvement) and SLND (4 lymph nodes); a radical hysterectomy (type C1) and systematic pelvic lymphadenectomy (26 lymph nodes) (because of ICG non-display of lymphatic pathways and high-risk disease). These two patients were not included in the statistical comparative calculations. Lymph node metastases were detected in 6 patients (5.1%) (2 MMRd, 2 NSPM, 1 p53mut., and 1 with unknown molecular subtype). Of note is that one patient had an NSMP EEC, G1, pT1a, and one positive sentinel (targeted) lymph node (without extracapsular spread). Other patients with positive lymph nodes had pT stages 1b and 2. Due to the performance of PMMR + TCL, additional histological findings were detected in the resected compartments (spaces between the uterus and lymph nodes) in 7 (8.8%) patients. These were: endometriosis detected in vascular mesometrium (lateral parametria) in 5 patients (2 patients with MMRd and 3 NSMP); endometriosis with focal transformation into endometrioid carcinoma (NSMP type) in one patient; deposits of carcinoma in the fatty-lymphatic tissue (removed ‘en block’) around negative targeted lymph nodes in another patient (p53mut., pT2, pN0). Thirteen (11%) patients experienced complications, of which 3 (2.5%) were classified as grade III: One intraoperative and two postoperative complications. Ten patients experienced postoperative grade I-II complications (Tables  4 , 5 and 6 ). Table 4 Surgical complications (all), Clavien-Dindo classification (all patients, n  = 118) Grade n (%) I 1 (0.8%) II 9 (7.6%) IIIa 1 (0.8%) IIIb 2 (1.7%) Surgical complications (all), Clavien-Dindo classification (all patients, n  = 118) Table 5 Surgical complications (intraoperative and 30-day postoperative) stratified into types of surgery ( n  = 116), Clavien-Dindo classification Procedure Variable PMMR + TCL ( n  = 79) rSH + SLND with afferent lymphatic vessels ( n  = 17) rSH + SLND only ( n  = 20) p Any complications n (%) Yes No 10 (13.0%) 69 (87.0%) 1 (5.0%) 19 (95.0%) 2 (12.0%) 15 (88.0%) NS* Intraoperative complications 1 (1.3%) † 0 0 - Postoperative (30 days) complications, grade I or II 8 (10.1%) 1 (5.9%) 1 (5.0%) - Postoperative (30 days) complications ≥III 1 (1.3%) 1 (5.9%) 0 - Footnote: * Chi 2 Pearson test, †, intended PMMR + TCL but not performed, because at an early stage of surgery, conversion to laparotomy (see details in the text) Abbreviations: NS  not statistically significant, PMMR  peritoneal mesometrial resection, SLND  sentinel lymph nodes dissection, TCL  targeted compartmental lymphadenectomy, rSH  robotic simple hysterectomy Surgical complications (intraoperative and 30-day postoperative) stratified into types of surgery ( n  = 116), Clavien-Dindo classification Any complications n (%) Yes No 10 (13.0%) 69 (87.0%) 1 (5.0%) 19 (95.0%) 2 (12.0%) 15 (88.0%) Footnote: * Chi 2 Pearson test, †, intended PMMR + TCL but not performed, because at an early stage of surgery, conversion to laparotomy (see details in the text) Abbreviations: NS  not statistically significant, PMMR  peritoneal mesometrial resection, SLND  sentinel lymph nodes dissection, TCL  targeted compartmental lymphadenectomy, rSH  robotic simple hysterectomy One grade IIIb complication happened intraoperatively. A left obturator nerve was cut at an early stage of performing ICG-marked left-side lymph-node mobilization in a 46-year-old patient with BMI = 20.5, Charlson Index of 0, who underwent two surgeries in the past – right hemicolectomy for Crohn’s disease and excision of the periovarian cyst. Before mobilization of the lymph nodes, all spaces were visualized, and a distal part of the obturator nerve was exposed. After completion of the mobilization of the left side lymph nodes, an obturator nerve injury was detected at the level and below the left common iliac vein bifurcation. There was no nerve reaction to coagulation (mono- or bipolar) at any time of preparation. Due to the retraction of the nerve, a conversion to laparotomy was immediately made. After obtaining full exposure to the retroperitoneal space, between the iliopsoas muscle and iliac vessels, the proximal part of the obturator nerve was mobilized, the edges with signs of coagulation at both ends were excised, and the nerve was sutured end-to-end with 5.0 polypropylene (Prolene ( ® ); Ethicon) suture. The lymph-node dissection was completed on the right side using ICG visualization via the DaVinci camera, followed by C1 radical hysterectomy. Postoperatively, the patient underwent a program of rehabilitation. Apart from the previous, the postoperative course was uneventful, and the patient was discharged home on the 10th postoperative day, walking without assistance. She required no adjuvant oncological treatment, and there was no long-term disability of the left lower extremity after the program of rehabilitation. Due to an intestinal obstruction sign, a laparotomy was performed (IIIb complication) in a patient on the 5th day after rSH with SLND and resection of afferent lymph vessels. A small bowel loop was incarcerated in the umbilical hernia (that existed before surgery for EC). No bowel necrosis was detected. A hernia was resected with the implantation of a net. A patient recovered well and was discharged home on the 6th postoperative day. Eleven patients experienced grade I-II (9.3%) complications. One patient was readmitted with an intestinal-like discharge from a vagina. Later, it self-diminished and stopped. Abdominal CT with bowel contrast did not reveal any signs of fistula. The patient did not develop any systemic or local infection. A small dehiscence in the vaginal apex healed spontaneously. She was discharged, and further follow-up was uneventful. Finally, we suspected a single episode of rectal-sigmoid diverticulum perforation (during the surgery, we saw bowel diverticula), evacuation of the content via vaginal dehiscence, and spontaneous closure. Another patient was readmitted because of vaginal waterish discharge on the 10th day after the PMMR + TCL. A vaginal apex dehiscence was detected. We did not reveal any signs of urinary fistula. Imaging did not show any abnormalities. The total (bilateral) number of lymph nodes removed with the TCL was 10. Lymphorrhea via vaginal dehiscence was diagnosed. The patient received prophylactic antibiotics. After a few days, the discharge diminished, and there were no signs of free fluid in the pelvis on the ultrasound. The vaginal apex was resutured in local anesthesia via vaginal approach (Clavien-Dindo IIIa). She was discharged, and a further 30-day follow-up was uneventful. However, she was readmitted on the 94th day after surgery with the next vaginal apex dehiscence, with the small bowel visualized in the vaginal apex (without intestinal incarceration). She had no clinical or laboratory symptoms of infection. She underwent laparoscopic visualization (no additional pathology found), vaginal apex (the dehiscence) was excised, and the apex was resutured with interrupted, PDS 2.0 sutures. Additionally, bladder and rectum peritoneal flaps were united over the vaginal stump. Of note is that the patient underwent a surgery for obesity (gastric sleeve) in the past (which might have an impact on the patient’s nutrition and, secondary a wound healing). Two patients were readmitted with pelvic pain. The imaging revealed a localized hematoma above the vaginal apex. One patient was diagnosed with hemophilia before. In one patient the hematoma evacuated spontaneously via small vaginal dehiscence. In the second patient, it was observed in imaging. None of the patients had progression of the hematoma. Both were managed conservatively. Other patients experienced symptoms of pelvic/abdominal pain and fever. The rehospitalization was deemed necessary. Peritoneal signs were negative. Clinical and laboratory results showed signs of infection; however, no abscess or any other focal lesions were detected on imaging in any of these patients. Empirical intravenous antibiotics were introduced, and it was effective in all these cases. After a few days of hospitalization, cessation of symptoms, and normalization of laboratory tests, patients were discharged home, and further follow-up was uneventful. There was no need for blood transfusion in any of the patients in the analyzed cohort. Table 6 A list of surgical complications (intraoperative and 30-days postoperative), according to procedure type, with management and the outcomes (all patients, n  = 118) Complication Clavien-Dindo classification Type of procedure (PMMR + TCL or rSH_SLND or other) Management Outcome Obturator nerve injury (intraoperative) IIIb Not possible to detemine * ( n  = 1) Immediate laparotomy, nerve mobilization and suturing, completion of surgery, followed by postoperative rehabilitation resolved Intestinal obstruction (postoperative) IIIb rSH + SLND ( n  = 1) Laparotomy, evacuation of the intestinal loop incarcerated in the umbilical hernia resolved Vaginal dehiscence with lymphorrhea IIIa PMMR + TCL ( n  = 1) Vaginal resuturing in local anaesthesia Repeated dehiscence (94th day) – vaginal resuturing via laparoscopy Pathological (intestinal-like) discharge from vagina (limited vaginal apex dehiscence) II rSH + SLND ( n  = 1) Diagnostics for intestio-vaginal fistula – negative, antibiotics IV and observation (intraoperative diagnosis of multiple recto-sigmoid diverticuli) resolved Hematoma over the vaginal apex II PMMR + TCL ( n  = 2) Antibiotics IV, LMH cessation, tranexamic acid, and observation resolved Pain and fever without radiologic evidence of abscess, hematoma etc. II PMMR + TCL ( n  = 6) rSH + SLND ( n  = 1) Antibiotics IV and observation resolved * Nerve injury happened at an early stage of the surgery, at mobilization and dissection of the sentinel lymph node, with early conversion to laparotomy Abbreviations: IV intravenous, LMH  low molecular heparin, PMMR  +  TCL  peritoneal mesometrial resection and targeted compartmental lymphadenectomy, rSH  +  SLND  robotic simple hysterectomy and sentinel lymph nodes dissection A list of surgical complications (intraoperative and 30-days postoperative), according to procedure type, with management and the outcomes (all patients, n  = 118) Not possible to detemine * ( n  = 1) rSH + SLND ( n  = 1) PMMR + TCL ( n  = 1) rSH + SLND ( n  = 1) Diagnostics for intestio-vaginal fistula – negative, antibiotics IV and observation (intraoperative diagnosis of multiple recto-sigmoid diverticuli) PMMR + TCL ( n  = 2) PMMR + TCL ( n  = 6) rSH + SLND ( n  = 1) * Nerve injury happened at an early stage of the surgery, at mobilization and dissection of the sentinel lymph node, with early conversion to laparotomy Abbreviations: IV intravenous, LMH  low molecular heparin, PMMR  +  TCL  peritoneal mesometrial resection and targeted compartmental lymphadenectomy, rSH  +  SLND  robotic simple hysterectomy and sentinel lymph nodes dissection We had the follow-up data available for all but 4 patients from our cohort, with the median time of 16.2 months (range: 2–29). Any adjuvant treatment was applied for 54 (47.4%) patients, with 37 (32.5%) receiving pelvic radiation (one refused), 24 (21.1%) vaginal brachytherapy, and 37 (32.4%) chemotherapy (one refused, and one had therapy discontinued due to chemotherapy-related complications after the first dose). No other treatment modalities were provided for the patients. In the follow-up period, all patients were alive; three were diagnosed with secondary malignancy: breast cancer (EC: pT1a pN0, NSMP ER+), colon cancer (EC: pT1a pN0, p53mut.), nodular lymphoma (EC: pT1a pN0, NSMP ER+).

Endometrial cancer (EC) is the most common gynecological malignancy in Europe [ 1 ]. Primary management is surgery (if not contraindicated). A simple hysterectomy (no parametrial resection is recommended, even in cases of cervical involvement) with bilateral salpingo-oophorectomy (ovarian preservation possible for defined premenopausal women) are standard procedures. Lymphadenectomy is generally considered a staging procedure [ 2 – 4 ]. A sentinel lymph node dissection (SLND) can result in equal oncological outcomes to systematic lymphadenectomy (as the primary staging modality) in node-positive patients receiving adjuvant treatment [ 5 ], even for high-risk histology [ 6 , 7 ]. Clinical, histological, and molecular factors guide decisions on adjuvant treatment [ 2 – 4 ]. Contrary, a compartmental resection of the uterus together with its lymph vascular system and first-line nodes “en bloc” (resecting the lymphatic network between tumor and first-line nodes) as performed in robotic Peritoneal MesoMetrial Resection and Targeted Compartmental Lymphadenectomy (PMMR + TCL) is an option for therapeutic purposes of the surgery [ 8 – 13 ]. This strategy follows the concept of the ‘cancer field surgery’, where the surgery can optimize locoregional control without adjuvant radiotherapy, without increasing morbidity [ 11 – 13 ]. This concept follows previously published data on cancer field surgery in cervical [ 14 ], vulvar [ 15 ], and rectal [ 16 ] cancers. This study aimed to evaluate the feasibility and safety of performing the PMMR + TCL for the surgical management of patients with EC, performed by a team newly trained in robotic surgery (DaVinci system), but already experienced in performing cancer field surgery for cervical cancer [ 17 ]. The primary outcome of interest was the intraoperative and 30-day postoperative complications rate after PMMR + TCL as compared to simple hysterectomy and SLND.

It was feasible for an experienced in cancer field surgery operator to perform the PMMR + TCL right after launching the DaVinci system approach for patients with EC. However, despite a “wish” to perform the PMMR + TCL in every patient, this was not the case. Several reasons contributed to this. First, the PMMR + TCL required more operative time than rSH + SLND, and generally, we were able to perform the procedure in younger patients with lower BMI and fewer co-morbidities. This was likely due to the intention of conducting shorter and less complex surgeries in patients at higher risk of surgical complications, such as high BMI, older age, significant co-morbidities, or poor tolerance to prolonged anesthesia. Second, in patients who had previously undergone pelvic surgeries with multiple adhesions or therapies (e.g., radiotherapy for rectal cancer) or who have coexisting diseases (e.g., endometriosis), it can be challenging or even impossible to perform PMMR + TCL correctly because of disrupted anatomical spaces/cancer fields of interest. However, the severity of anatomical changes due to previous surgeries cannot be predicted until the beginning of the surgery, because every case may differ in that matter, even after the same procedure. Third, it would probably be more justified to aim at performing PMMR + TCL even in “poorer conditions” if it were regarded as a therapeutic procedure by all oncologists at multi-disciplinary team meetings. However, despite being shown as an effective therapeutic procedure [ 12 , 13 ] and ongoing trial [ NCT04504006 ], it has not yet been included in general guidelines [ 3 , 26 ]. Moreover, a “new” parading with different oncological outcomes and treatment regimens for different molecular subgroups seems to introduce even more complexity into oncological management and adds bias regarding the significance of surgery in EC treatment, with surgery being considered merely a staging procedure [ 2 – 4 ]. This contrasts with the cancer field surgery concept [ 12 , 13 , 18 ]. Major surgical complications occurred in 3 patients. One intraoperative obturator nerve injury at the early stages of surgery that needed conversion to laparotomy and nerve reconstruction, one postoperative bowel obstruction after rSH + SLND with afferent lymph vessels that needed laparotomy (incarcerated in the umbilical hernia small bowel), and one vaginal dehiscence requiring vaginal sutures. Grade I-II complications tended to be more often after the PMMR + TCL than rSH with SLND +/- resection of afferent lymphatic vessels. The most significant intraoperative complication in our first patients operated on a robotic approach was the obturator nerve cutting. It happened despite the patient not being considered at high risk for surgical morbidity, although she underwent two surgeries in the past; and despite visualization of the obturator nerve between the external iliac vessels and the umbilical artery (video: https://youtu.be/c03VH-Va7VE ) before lymph node dissection. Moreover, the main surgeon was experienced in gynecological oncology surgery, including performance of the TMMR with tLND [ 17 ], and significantly contributed to the Department’s ESGO (European Society of Gynecological Oncology) certificate for surgery for advanced ovarian cancer [ 27 ] (obtained in 2020). However, the experience was in open surgery. Thus, the only explanation for the complication was the change to the robotic approach, with its different visualization and perspective. Subsequently, all patients with ICG-marked obturator lymph nodes had an obturator nerve separated from the fatty-lymphatic tissue along its length from below the iliac vessels’ bifurcation and distally. The complication cannot be directly linked to any type of surgery (PMMR + TCL or rSH + SLND) because the obturator nerve injury happened at an early stage of sentinel/targeted lymph node mobilization from the obturator fossa. Importantly, it is essential to visualize and remove all lymph nodes at risk, both those visualized by ICG and if not, the nodes from localizations at risk [ 8 , 28 , 29 ]. Thus, “difficult” localization cannot be a “contraindication” for the removal of sentinel/suspicious lymph nodes. Burbano-Luna J et al. collected eight studies, all case reports on obturator nerve injury during laparoscopic lymphadenectomy. Most ( n = 6) were recognized during the surgery; three were repaired with end-to-end anastomosis; three were reconstructed using sural nerve grafting; one was managed with neurolysis and end-to-end anastomosis [ 30 ]. In a total of 3558 pelvic lymphadenectomies added to prostatectomy, the incidence of obturator nerve injury during laparoscopic and robot-assisted procedures was 3 (0.1%) and 2 (0.1%), respectively. Simultaneous repair was performed successfully in all cases, and there were no complications associated with obturator nerve function during long-term follow-up [ 31 ]. In the literature review, the proximal part of the obturator nerve was at the highest risk for injury during pelvic lymphadenectomy, representing 77.8% of reported cases of obturator nerve injury [ 31 ]. This was the localization of the injury in our patient. Buderath et al. compared surgical morbidity after the PMMR + TCL and the total laparoscopic hysterectomy (TLH)+\- systematic lymphadenectomy (LND) performed for EC and after the TLH performed for benign conditions. Among their patients’ cohort, intraoperative complications appeared in 2.0%, 0%, and 2.7%, respectively. Postoperative complications (all grades) were noted in 13.7%, 10.0 and 7.6%, respectively, of which 3.9%, 5.0%, and 1.9% were classified as grade ≥III Clavien-Dindo classification [ 12 ]. In the same institution, according to the most recent data, among 135 patients who underwent PMMR + TCL for EC, 6 (5.9%) patients experienced intraoperative complications, of which 5 were serosal bowel lesions, and one transverse colon perforation. Postoperative complications were diagnosed in 18.5% of patients (25/135), of which 5 and 2 patients experienced grade III and IV/V complications, respectively. Voiding disorders, infections, and thromboembolic events were the most frequent grade III complications. Among grade IV/V, there was one bowel lesion (reoperations required) and one pulmonary embolism (with intra-abdominal bleeding secondary to lysis therapy). Of note is that the mean BMI in this cohort was 33.8 kg/m 2 (16.7–83.7). This might have an impact on the risk of bowel lesions, although the authors did not comment on this issue [ 13 ]. In our cohort, we did not report intraoperative bowel lesions. However, there was one patient who was re-admitted with a bowel-like discharge from a vaginal dehiscence. This was a self-limiting discharge, thus considered probably as a perforation of the colon diverticulum with its self-closure. A rate of grade III (Clavien-Dindo classification)/major postoperative complications after robotic-assisted surgery for EC was reported as 2.6% − 22% patients [ 32 – 34 ]. The rate of minor complications was reported on 5.5% − 16.6% for robotic-assisted surgery [ 34 , 35 ]. This is comparable to our series (2.5% of grade III intra- and postoperative complications; 8.5% grade < III postoperative complications), which is reassuring, especially given that we were a newly trained team in robotic approach and that the majority of patients underwent PMMR + TCL. Casarin et al. reported that no significant differences in intra-operative and post-operative complications, re-admissions, and re-operations were noted between patients who underwent robotic-assisted SLND versus systematic pelvic lymphadenectomy or hysterectomy alone for apparent early-stage EC ( n = 621). Patients who underwent hysterectomy alone vs. hysterectomy with SLND experienced intraoperative complications in 1.7% and 0.5%, respectively. For 30-day postoperative grade II/III complications, it was 8.1% and 5.3% for hysterectomy alone vs. hysterectomy with SLND, respectively [ 34 ]. In a meta-analysis including 769 patients with EC, systematic LND resulted in a significantly higher risk of surgical complications relative to SLND for the laparoscopic approach (RR 2.10 [95% CI 1.37 to 3.21] and was comparable to the robotic approach [ 36 ]. In prospective trials (performed by experienced surgeons in robotic procedures) on assessing the diagnostic accuracy of a pelvic sentinel lymph node algorithm in high-risk EC, there were no intraoperative complications associated with the SLND per se [ 7 , 37 ]. For T2 EC, the performance of radical vs. simple hysterectomy did not provide a significant survival benefit for either overall or progression-free survival [ 38 , 39 ], thus, international guidelines recommend against performing radical hysterectomy for patients with EC, unless it would enable R0 resection [ 2 , 3 , 26 ]. Contrary to the above discussion, the PMMR + TCL [ 9 , 13 ] is considered a surgical single (without the need for irradiation) treatment modality for loco-regional control of EC. It follows principles of the cancer field surgery concept, where the whole embryologically defined anatomical field, together with the corresponding lymphatic system, without its disruption, is resected. Interestingly, in our series of PMMT + TCL specimens, pathologists found additional lesions (endometriosis alone or with focal transformation into carcinoma or deposits of carcinoma in the fatty-lymphatic tissue around negative targeted lymph nodes) exactly in the ‘en bloc’ resected fields. From the first study on PMMR + TCL efficacy, the low loco-regional recurrence rate of 2.9% was reported despite a very low percentage of adjuvant radiotherapy and the 56.0% incidence of intermediate/high-risk tumors [ 10 ]. According to the most recent data ( n = 135) from the same institution, even though 50.4% of patients had an indication for postoperative radiotherapy (following international guidelines), the rate of postoperative irradiation administered was 10.4%, and the overall recurrence rate was 8.1%, with 1.5% ( n = 2) isolated locoregional recurrences [ 13 ]. The lymph from the uterus is drained via three lymphatic pathways: lower pelvic (pathway I), upper pelvic (II), and mesonephric direct paraaortic (III) [ 8 ]. However, the true sentinel nodes are found in EC in pathways II and III only if ICG is injected in the corpus uteri. According to functional studies with ICG injections in the vagina, uterine cervix, and uterine corpus, the lower pelvic pathway is involved in cervical cancer only [ 8 ]. On the contrary, Persson et al. reported that in women with non-endometrioid histology, isolated metastases along the lower pelvic pathway were identified in 2/53 (3.8%) of node-positive women [ 28 ]. However, the Authors did not provide details of whether cervical involvement was present in these patients. We did not analyze these details in our series. Probably, one should explore the lower pelvic pathway in T2 ECs. According to Kimmig et al. [ 8 ], the first branching of the collector usually occurs before the sentinel nodes, so regularly, there are two true sentinels on each side (in the upper pathway). The first branch runs medially, ventrally of the external iliac vessels. The second branch runs medially, dorsally around the external iliac vein, further branching to the lateral iliac lymph basin, but always ventrally to the obturator nerve [ 8 ]. This implies the need to explore the obturator fossa always. Thus, the mean number of lymph nodes in the TCL was 8.8 (2–28) [ 13 ]. In the concept of separate SLND, a median of 4 (range 1–8) ICG-defined nodes were retrieved along the upper pathway [ 28 ]. In our series, the median number of nodes in the TCL was 8 (both sides). A compartmental resection (TCL) of first-line nodes can be performed in EC without reducing the safety of detection of node metastases [ 8 ]. In a different but to some extent comparable concept, selective removal of lymph nodes at typical pelvic positions could replace a full side-specific pelvic lymphadenectomy when sentinel mapping is unsuccessful [ 28 ]. However, there is a significant difference between the two concepts. In the PMMR + TCL, one removes the uterus and all close corresponding cancer fields, including lymph vessels, intercalated lymph nodes in vascular mesometria [ 40 ], and targeted compartmental lymph nodes, without cutting lymph vessels in between. Thus, it is a “clear” oncological procedure with a minimum risk of cancer cell spillage [ 11 ]. The clinical effect with excellent loco-regional control after the cancer field surgery for EC (PMMR + TCL alone) was shown, although 40.0% would be candidates for adjuvant irradiation [ 13 ]. On the contrary, when performing a selective removal of lymph nodes, one cuts lymph vessels [ 28 ], which poses a risk of cancer cell spillage if present in the lymph vessels. The disadvantage of PMMR + TCL is that it takes a longer surgical time (mean skin-to-skin time 189 min [ 13 ]., 213 min. in our cohort), as compared to simple hysterectomy for EC (mean 129.6 min [ 12 ]). or to selective removal of lymph nodes at typical proximal obturator and interiliac positions (68 min [ 28 ]). and simple hysterectomy (71 min [ 28 ]). (in our cohort rSH + SLND – 192.5 min.). The time of surgery and the risk of surgical complications must be addressed when scheduling fragile patients with co-morbidities. The strength of the study is that it provides information on the feasibility and safety of the cancer field surgery for EC patients being carried out on consecutive patients treated in a cancer center, outside and without direct support of the designers of the procedure. It may also be of interest to centers transforming to minimally invasive cancer field surgery. We tried to be clear and transparent with definitions and procedural steps, including the video, to enhance the reproducibility. The detailed pathological report was available for all patients, which enabled showing some additional findings in the resected cancer fields, which may support the oncologic rationale of the procedure. Lastly, we added information on molecular profiling of the EC, which has never been done before, along with the subject of PMMR + TCL. There are a few limitations of our study. First, it was a retrospective, single-institution experience with a robotic approach for surgery in patients with EC. A team was newly trained, without any prior experience in other robotic-assisted gynecological procedures. This may have an impact on the feasibility of performance of the PMMR + TCL as well as the complication rate. On the other hand, the team was experienced in gynecologic oncology surgery as well as the cancer field surgery concept. Second, there was no pre-selection for any type of surgery, but rather it was an observational study on how often, and on what type of patients, we were able to perform the PMMR + TCL, and what factors prevented us from finalizing the procedure. So, the selection was on every consecutive patient, with all factors considered, like in the “real-world” scenario. It is even more important as a “typical patient” with EC is obese, has co-morbidities, and had undergone some prior surgeries in the past (29–55% in our series), of which cholecystectomy seems the most frequent one. Results proved that we tended to perform simpler and quicker procedures – the rSH + SLND (instead of PMMR + TCL) in older, more obese, and co-morbid patients. The fact of undergoing surgery in the past was not statistically different between groups. However, “the previous surgery” is such a multifactorial issue, with additional individual predisposition to create adhesions, that it should not be considered in the decision-making on the type of procedure until the very start of the surgery and real evaluation of the severity of adhesions. Third, patients were described with limited clinical data, and we did not provide a multivariate analysis on factors with a possible impact on complication rate. However, the number of events of interest (complications) was low, with a limited number of cases in each of the 3 groups (different procedures), so we did not perform a multivariate analysis, because it could provide unreliable results. Additional clinical factors, like multiple adhesions after prior surgeries, prior pelvic radiotherapy, or deep infiltrating endometriosis, made it impossible to perform PMMR + TCL as described by Kimmig et al. Acknowledging above mentioned limitations is important for the generalizability of our results for patient selection, as well as to avoid overinterpreting complication rate differences. Not exploring the full potential of the PMMR + TCL, namely avoidance of adjuvant radiotherapy (if indicated by guidelines), and providing the oncological outcomes of the therapeutic intent of the procedure, could be considered a next limitation of our study. However, at present, in the era of suggested molecular subtype-guided adjuvant treatment, we had no strong rationale to establish different management for those who underwent different procedures (PMMR + TCL). Even in the most recent study on PMMR + TLC [ 13 ], the molecular profiling of the disease was not considered. The patients were recruited between 2016 and 2023, so before the wider adoption of molecular profiling. For the present time, with molecular sub-typing of the disease, any comparisons of oncological outcomes may pose a challenge. However, in the view of many recent changes in understanding [ 41 ], staging [ 42 ], and management of endometrial cancer [ 2 , 3 , 26 ], it is rather a global problem. Molecular profiling is shifting disease stage and risk classification significantly, as compared to FIGO 2009, even 23–48% patients could be retrospectively upstaged [ 43 ]. Even major international guidelines, the European (ESGO-ESTRO-ESP) [ 26 ] and American (NCCN) [ 44 ], still do not provide coherent statements. The first stratifies risk groups and management based on molecular profiling, while the latter one guides adjuvant treatment based on more traditional clinical and pathological data. Some clinicians are even creating something in between, like the Betella algorithm [ 45 ]. The application of this algorithm appears safe for the patients while rationalizing resources that could thus be allocated otherwise. The application of the algorithm has one fundamental precondition: staging surgery must be properly executed by surgeons with the appropriate expertise [ 46 ]. The comparison between American and European protocols revealed some relevant disparities, and this could possibly cause differences in interpreting and applying protocols in clinical practice in small centers, leading to a lack of adherence to guidelines or even prompting a confusing mix of them [ 47 ]. A merit counterpoint to the FIGO 2023 staging system is provided in the literature [ 48 ]. For the mentioned reasons, we used the TNM staging system. Moreover, in all our patients, the adjuvant treatment was guided by European recommendations with molecular sub-typing of the disease (not by the cancer field surgery concept, even for those who underwent PMMR + TCL). It was because we did not want to provide more bias by applying a “new” surgical procedure to an already complicated system. However, now that we know the feasibility, safety, and are aware of challenges in understanding international guidelines, we could join the ongoing study on the PMMR + TCL ( NCT04504006 ) or suggest some amendments to the protocol, with the incorporation of molecular data into the analysis of oncological outcomes. Considering the surgery for EC as a staging procedure only, currently there are no established modifications of its extent based on molecular subtype, but the ovarian preservation (if needed), and the need to achieve R0 resection in advanced disease [ 26 ]. A combination of comprehensive surgical-pathological staging and molecular profiling followed by standard adjuvant treatment/follow-up is now being explored for the oncological outcomes in the EUGENIE trial [ 49 ]. Results may provide an input on the extent of the surgery needed for different molecular EC subtypes (e.g., without lymph node staging or the opposite, with additional surgical staging procedures required). Contrary, the PMMR + TCL belongs to the cancer field surgery, where the intent is treatment, with loco-regional control of the disease without the need for other means (irradiation) to achieve the control [ 13 , 18 ]. In this concept, the aim is to resect all the local compartments where the cancer cells may spread, together with regional lymph nodes and collector lymphatic vessels. The type of the surgical resection is determined by anatomical disease stage (ontogenetic staging) – PMMR for uterine-confined disease and the extended PMMR if the tumor involves deep cervical stroma [ 18 ]. Nevertheless, it would be worth exploring whether patients with a more favorable molecular disease profile would have comparable oncological results with intracompartmental resection (e.g., simple hysterectomy +/- SLND, without the need to resect vascular mesometria (parametria)) as compared to those after the PMMR + TCL. Of the most interest would be analyzing the loco-regional control of the EC after cancer field surgery for those with high-risk disease defined by molecular profile only (e.g., p53mut., uterus-confined disease). These patients could benefit from being spared from adjuvant irradiation (both pelvic and vaginal) and its implications for quality of life. Interestingly, the cancer field surgery applied to vulvar cancers provided no significant differences in clinical outcomes when stratified on two different molecular subtypes [ 50 ].

It was feasible and safe to perform the PMMR + TCL for patients with endometrial cancer by a newly trained team in the robotic DaVinci system, who had experience in the cancer field surgery. Meaningful additional histological findings could be detected in the resected compartments. The future research should explore the true benefits of no need for adjuvant radiotherapy for loco-regional disease control after the PMMR + TCL with stratification on EC molecular subtypes.

Supplementary Material 1. Description of surgical procedures details. Supplementary Material 1. Description of surgical procedures details.