Upfront radical hysterectomy is suitable for early-stage invasive cervical cancers, with a degree of parametrial resection that depends on histopathological prognostic factors (depth of infiltration, presence of lymphovascular involvement). The surgery of the uterus is associated with pelvic lymph node dissection (PLND), including external iliac, inter-arteriovenous, and obturator lymph nodes ^[1][2][3][4][8,13,14,15]. The procedure of a sentinel lymph node (SLN) biopsy is being increasingly used, for ultra-staging but also, given the significant morbidities reported in PLND, such as hemorrhaging, ureteral and/or nerve lesions, as well as lymphoedema ^[5][6][16,17]. The surgery of the uterus is recommended after ruling out, intraoperatively, the pelvic lymph node extension. The optimal surgical approach remains debatable. A randomized study showed that minimally invasive surgery was associated with more frequent disease relapses, compared to laparotomy-based hysterectomies (3-year disease-free survival of 91.2% vs. 97.1%, HR 3.74, 95% CI, 1.63 to 8.58) ^[7][18]. In a large retrospective cohort study, the deleterious impact of mini-invasive surgery was observed among the subgroup of patients with a tumor size > 2 cm (HR 2.31; 95% CI 1.37–3.90) ^[8][19].

Around 25–40% of stage IB cervical cancers present risk factors identified on the hysterectomy pathology specimen and required adjuvant treatment ^[9][20]. Patients with one or more adverse histopathological factors, such as positive or close surgical margins, positive lymph nodes, or microscopic parametrial involvement are at a high risk of relapse. Several retrospective series studied the long-term results of patients operated for early-stage cervical cancers without adjuvant treatments.

The most significant prognostic factors for stage IB cervical cancer are occult pelvic lymph node involvement found at time of surgery, and tumor size ^[10][11][12][21,22,23]. After a radical hysterectomy, the 5-year OS rate for patients with tumors ≤ 2 cm was 94–97%, compared with 69.9–80% of patients with tumors > 4 cm ^[11][22]. One randomized phase III study compared adjuvant chemoradiotherapy versus adjuvant pelvic radiotherapy in 268 patients. They had FIGO IA2-IIA cervical cancer, with positive pelvic lymph nodes and/or positive margins, and/or microscopic involvement of parametrium after an upfront radical hysterectomy and PLND ^[13][24]. Chemotherapy consisted of 3 cycles of cisplatin, plus fluorouracil every 3 weeks, with the first and second cycles given concurrently with the radiotherapy, which is 49.3 Gy in 29 fractions to the pelvic field. The 4-year progression-free survival (PFS) and 4-year OS were both significantly improved in the adjuvant chemoradiotherapy group compared with the adjuvant radiotherapy group (80% vs. 63% and 81% vs. 71%, respectively). The addition of concurrent chemotherapy to adjuvant radiotherapy is therefore considered to be the standard to improve survival of high-risk, early-stage cervical cancer patients ^[14][9].

Along with the abovementioned high-risk factors, tumor dimensions, deep stromal invasion, and lymphovascular space invasion, are intermediate-risk factors increasing the risk of relapse when combined ^[14][15][9,25]. A phase III randomized Gynecologic Oncology Group (GOG) trial including 277 patients with at least two intermediate-risk factors, compared adjuvant pelvic radiotherapy with a radical hysterectomy plus PLND alone. The planned pelvic dose was from 46 Gy in 23 fractions to 50.4 Gy in 28 fractions. A significant increase in 10-year PFS was observed in the group of patients who received post-operative radiotherapy, but the 10-year OS were similar in two groups ^[16][17][26,27]. This randomized clinical trial showed the benefits of adjuvant pelvic radiotherapy in these patients; however, MRI was not routinely used in the 1990s to evaluate the disease extension in the pelvis, and 18-fluorodeoxyglucose positron emission tomography/computed tomography (18-FDG PET/CT) only came into regular use a decade later. The disease staging is much more precise with modern imaging methods, possibly leading to fewer indications for adjuvant treatment through a better selection of patients for upfront surgery ^[18][19][4,6]. In the GOG trial published by Sedlis et al. in 1999, the criteria for adjuvant radiotherapy were defined for squamous cell carcinoma at an estimated recurrence risk of at least 30%, which is a very high threshold ^[13][24]. More recently, an ancillary analysis of GOG 49, 92, and 141 trials was conducted to better identify the risk of relapse according to tumor histopathological characteristics. Authors proposed histology specific nomograms that better represented the risk of relapse to guide adjuvant radiotherapy. Though the optimal threshold to indicate adjuvant radiotherapy is still unclear, they showed that risk factors for recurrence differed for squamous cell carcinoma and adenocarcinoma of the cervix, with tumor size being the risk factor associated with the highest risk for relapse in adenocarcinoma, and for squamous cell carcinoma, depth of invasion is the most important risk factor ^[15][25]. The place of adjuvant chemoradiation in patients with a combination of intermediate-risk factors is still under investigation.

Brachytherapy boost is frequently added in the adjuvant setting to the external radiotherapy to treat the vaginal cuff, where most locoregional relapses occur in patients treated with primary surgery. The benefit of adding vaginal brachytherapy was not clearly demonstrated ^[20][28], but retrospective series suggested that a treatment combining adjuvant intensity-modulated radiation therapy (IMRT) and a brachytherapy boost after radical hysterectomies in patients with cervical cancer, was associated with excellent long-term outcomes and limited rates of non-hematologic toxicity in patients with FIGO IB1-IIIC2 cervical cancers ^[21][29]. A retrospective study compared the treatment effect in 480 patients with at least two intermediate-risk factors, who received external beam radiotherapy with or without vaginal brachytherapy ^[20][28]. Patients who received brachytherapy had worse prognostic factors. The 5-year OS, 5-year local recurrence-free survival, and 5-year distant metastasis-free survival were all similar in both groups. Scarce data supported the use of vaginal vault brachytherapy in patients with positive surgical margins ^[20][28]. In selected situations, adjuvant vaginal brachytherapy was used as an exclusive adjuvant treatment in patients with pT1a-1b, pN0 cervical cancer and intermediate histopathological risk factors (LVSI) after a radical colpohysterectomy, in order to avoid the morbidity of adjuvant EBRT. In the only published series testing exclusive adjuvant brachytherapy, 10% of patients experienced relapse all having peritoneal cavity or lymph node failure. No vaginal or isolated pelvic nodal failure ^[22][30].

In a randomized study including patients with stage IB and IIA cervical carcinoma, radical surgery and radiotherapy showed similar survival outcomes, but patients in the surgery arm had a higher proportion of treatment-related adverse events ^[23][24][31,32]. Two-thirds of the patients in the radical surgery arm required adjuvant radiotherapies due to pejorative prognostic factors (stage T2b or greater disease, less than 3 mm of uninvolved cervical stroma, positive margin and lymph node metastases). In approximately one half of the cases, the indication was based on local risk factors, such as deep stromal invasion, positive margins, parametrial extension. Adjuvant radiotherapy increased the risk of post-treatment complications. One major difficulty of post-operative radiotherapy in patients who had their uterus removed is that gastrointestinal structures, especially the bowel, are more affected than in upfront radiotherapy ^[23][31]; therefore, patients with tumors that are close to 35–40 mm, or with pejorative factors identified from conization, may be preferentially treated with upfront chemoradiation plus brachytherapy. When there are indications that postoperative radiotherapy will be necessary, chemoradiation plus brachytherapy should be preferred to avoid cumulative morbidities of surgery and adjuvant treatment. Radiotherapy is also indicated for treatment of patients with medical contraindication for surgery, followed by a brachytherapy boost.

Adjuvant postoperative and upfront radiotherapy cause long-term gastrointestinal tract symptoms in patients treated with pelvis radiotherapy or brachytherapy for gynecological malignancies; the risk of long-term severe gastrointestinal sequelae still persists for a long time after radiotherapy ^{[13][16][23][25][26][27][28][29][30]}[24,26,31,33,34,35,36,37,38]. Two phase II studies demonstrated that post-operative IMRT to the pelvis in endometrial carcinomas caused an acceptable rate of short-term bowel adverse events and gastro-intestinal toxicities ^[31][32][39,40]. The randomized phase III NRG 1203 trial including 278 patients with gynecologic malignancies compared patient-reported acute toxicity and quality of life using a standard outcome questionnaire after the post-operative standard pelvic 3D-CRT and IMRT. Pelvic IMRT was associated with significantly less gastrointestinal and urinary tract toxicities compared to 3D-CRT at week 5 and 1 year ^[33][41]. A recent randomized phase III PARCER trial including 300 patients with cervical cancers, also compared late gastrointestinal tract toxicities after post-operative IMRT with daily image-guidance compared with 3D-CRT. Seventy-seven percent of the patients also received concurrent chemotherapy. IG-IMRT was associated with a significantly lower 3-year incidence of grade ≥ 2 late gastrointestinal tract toxicities compared with 3D-CRT (21.1% vs. 42.4%, HR, 0.46; 95% CI, 0.29 to 0.73; p < 0.001), as well as a 3-year incidence of grade ≥ 2 with any late toxicities (28.1% vs. 48.9% HR, 0.50; 95% CI, 0.33 to 0.76; p < 0.001). No difference in the 3-year disease control was observed, confirming the safety of IMRT technique ^[34][42]. Evidences from these two randomized phase III studies established the superiority of IMRT as adjuvant radiotherapy in gynecological cancers to decrease treatment-related morbidities, without jeopardizing treatment efficacy.

In order to reduce the treatment-related risk of combining surgery and adjuvant radiotherapy, neo-adjuvant brachytherapy in early-stage is propsoed in a few expert centers to patients with unfavorable histoprognostic factors. Retrospective series have reported on the efficiency and safety of neo-adjuvant brachytherapy, which consists of performing intracavitary brachytherapy 6 to 8 weeks before surgery. Rretrospective studies including patients with squamous cell carcinoma and adenocarcinoma suggested that neo-adjuvant brachytherapy is an attractive option for patients with aforementioned local pejorative prognostic factors such as lympho-vascular invasion or a tumor size > 2 cm. Neo-adjuvant brachytherapy is a valid option according to the latest European guidelines for cervical cancer management in centers familiar with this approach ^[14][35][9,43]. The brachytherapy delivers a considerable dose of radiation to the cervix uteri, the proximal part of the parametrium, and upper third of the vagina ^[35][43]. Due to the very sharp dose gradient of brachytherapy, a dose received by at risk organs in the vicinity, especially the bowel, is usually low. In retrospective data, 70% of patients have achieved a complete histological response by the time of surgery ^[36][44]. The 5-year estimated OS was 84.5%, and the 5-year DFS was 84.4%, with minimal treatment-related complications. In addition, fewer than 1% of patients have residual tumor cells in the parametrium, suggesting that extra-fascial hysterectomy could be an alternative to colpohysterectomy in this context ^[36][44]. A retrospective study compared upfront surgery with adjuvant radiation therapy (external beam radiation and/or vaginal brachytherapy) or neo-adjuvant radiotherapy (mainly uterovaginal brachytherapy) followed by surgery. This series showed that more frequent post-operative acute ureteral complications were observed in patients with neo-adjuvant radiotherapy (2.3% vs. 0.6%), but 10-year grade 3 and 4 late treatment-related complications were three times more frequent among patients treated with adjuvant external radiation compared with patients treated with neo-adjuvant brachytherapy (22% vs. 7%) ^[37][45].

With the development of the sentinel lymph node (SLN) biopsy ^[38][7], and pathological ultra-staging of SLN techniques, new questions have emerged, including the strategy to adopt the adjuvant treatment when micrometastases (lymph node metastases > 0.2 mm and up to 2 mm) or isolated tumor cells (ITCs) (tumor cells clusters < 0.2 mm) emerge. By doing multiple serial sectioning with immunohistochemistry staining, pathologic ultra-staging increases the detection rate of low-volume disease ^[39][46]. In 4–15% of SLN biopsy samples from early-stage cervical cancer patients, micrometastases were identified ^[40][41][47,48]. An international retrospective study examined the significance of micrometastases and isolated tumor cells for the disease prognosis in 645 patients treated for FIGO stage IA-IIB cervical cancers ^[42][49]. Macrometastases, micrometastases, and isolated tumor cells were detected by ultra-staging combined with pelvic non-SLN in 21.1%, 7.1%, and 3.9% of patients, respectively ^[42][49]. In this study, 85.3% of patients with macrometastases, 82.6% with micrometastases, 52% with isolated tumor cells, and 10.5% with negative pelvic nodes, received adjuvant therapy. The presence of isolated tumor cells was not a prognostic factor for recurrence free survival or overall survival; however, the presence of macrometastases or micrometastases was significantly associated with reduced overall survival, which was comparable to that of patients with macrometastases ^[42][49]. The impact of micrometastases and isolated tumor cells on survival was also examined in SLN samples from 139 cervical cancer patients with FIGO IA2-IB1 tumors treated in the SENTICOL I study ^[43][50]. In this study, the presence of the micrometastatses was not a prognostic factor for disease-free survival, possibly due to the lack of statistical power ^[43][50]. The adaptation of adjuvant treatment strategies should probably consider the presence of micrometastases, especially for indication of adjuvant radiotherapy, but not the presence of isolated tumor cells in the SLN biopsy.

There is currently no consensus concerning the role of sequential chemotherapy in addition to postoperative radiation in patients with early-stage cervical cancer who show adverse pathological factors. The sequence of chemotherapy using paclitaxel–carboplatin every 3 weeks, followed by radiotherapy, did not improve 2-year PFS or 5-year OS compared with the concomitant chemoradiotherapy using weekly cisplatin (81.8% and 87.2%, p = 0.235) in a phase III randomized trial including 263 patients; however, the sequential chemotherapy had lower hematological toxicities, but a much higher neurotoxicity and alopecia rate ^[44][51]. The recent phase III STARS clinical trial, including 1048 patients with FIGO IB to IIA tumors, showed that adjuvant sequential chemoradiotherapy using paclitaxel–cisplatin every 3 weeks was associated with a higher rate of 3 year DFS compared with radiotherapy alone (90% vs. 82%, HR, 0.52, 95% CI 0.35–0.76), or to concomitant chemoradiotherapy (90% vs. 85%, HR, 0.65, 95% CI 0.44–0.96), and a higher 5 year OS compared with adjuvant radiotherapy alone (92% vs. 88%, HR, 0.58; 95% CI 0.35–0.95) ^[45][52]. This study had some limitations, but it suggests a potential benefit of adjuvant sequential chemoradiation in patients with high-risk features.