Outcomes of node-positive cervical cancer treated with neoadjuvant chemotherapy followed by radical surgery: a single-center approach to a challenging disease

Abstract

Objective

Chemoradiotherapy with brachytherapy (CCRT-B) is the standard treatment for locally advanced cervical cancer (LACC), but prognosis remains poor for patients with lymph node involvement (International Federation of Gynecology and Obstetrics [FIGO] stage IIIC1-2) and bulky disease. This study aimed to evaluate recurrence patterns, salvage therapies, and survival outcomes in patients with FIGO IIIC1-2 cervical cancer (CC) treated with platinum-based neoadjuvant chemotherapy (NACT) followed by radical surgery (RS), and to explore NACT + RS as a potential alternative to CCRT-B.

Methods

This retrospective study included 97 patients with FIGO IIIC1–2 CC treated between 2012 and 2022 at Policlinico Umberto I, Rome. All patients received 3 cycles of platinum-based NACT followed by radical hysterectomy, bilateral salpingo-oophorectomy, and systematic pelvic lymphadenectomy, with para-aortic lymphadenectomy performed if indicated. Adjuvant radiotherapy was given in cases of positive vaginal or parametrial margins; chemotherapy or chemoradiotherapy was administered in patients with lymph node positivity after surgery.

Results

With a median follow-up of 51 months, 33% of patients died from the disease, 5% from other causes, 58% were alive without disease, and 4% were alive with disease. The 2- and 5-year overall survival (OS) rates were 80.1% and 61.1%, respectively. Progression-free survival was 69.5% at 2 years and 58.5% at 5 years. Median OS after recurrence was 1.01 years. Recurrences occurred primarily in pelvic lymph nodes (22%) and central pelvis (9%).

Conclusions

NACT followed by RS offers similar survival outcomes to CCRT-B in node-positive LACC and may be a viable alternative in specialized and well surgical-trained centers.

Trial Registration

ClinicalTrials.gov Identifier: NCT06752135

Synopsis

Neoadjuvant chemotherapy followed by surgery is effective for high-risk cervical cancer management. Survival outcomes are comparable to those of standard chemoradiotherapy treatment. Avoiding radiotherapy may reduce long-term side effects and improve quality of life. More personalized approaches should be explored in this setting.

INTRODUCTION

Cervical cancer (CC) is the second most diagnosed cancer and the fourth leading cause of cancer death in women, with an estimated 604,000 new cases and 342,000 deaths worldwide in 2020 [1]. In developed nations, elevated socioeconomic status and the implementation of extensive primary and secondary prevention strategies, such as human papillomavirus vaccination, have contributed to a notable decline in the incidence and mortality rates of CC over the past decades [2]. However, a substantial proportion of patients (about 40%–50%) continue to be diagnosed with locally advanced cervical cancer (LACC) at the time of clinical presentation (International Federation of Gynecology and Obstetrics [FIGO] stage IB2-IVA), with nodal involvement being the most important factor affecting prognosis [3].

Indeed, in patients presenting with LACC, nodal involvement occurs in 20% to 60% of cases depending on the FIGO stage, lymph-vascular space invasion and tumor diameter at the time of diagnosis [4,5]. Worldwide, national guidelines established concomitant chemoradiotherapy with brachytherapy (CCRT-B) as the standard treatment for patients diagnosed with LACC [3] based on observational studies [6,7]. Conversely, neoadjuvant chemotherapy (NACT) plus radical surgery (RS) has been extensively used in specialized centers and many prospective studies have been published with encouraging results in terms of survival data and complication rate [8], even if recent randomized trials did not confirm the superiority of such an approach [9,10].

After CCRT-B, patients still present high risk of persistent or recurrent disease (more than 40%) [11], especially in case of nodal metastasis, mainly with distant pattern of relapse [12,13]. These patients are often not suitable of curative treatment and present poor survival outcome (median survival rate after recurrence 7–10 months) in pre-immunotherapy/antiangiogenic era [14].

In this retrospective observational study, we would describe the pattern of recurrence, subsequent treatment and survival data of patients diagnosed with high risk LACC, i.e., patients with positive nodes (FIGO IIIC1-2) submitted to NACT + RS.

MATERIALS AND METHODS

1. Ethics statement

The local ethic committee approved the study (Azienda Policlinico Umberto I, protocol 0330/2024, rif. 7350). The study was registered in ClinicalTrials.gov, number protocol NCT06752135.

2. Data collection

Data were extrapolated by a prospectively collected institutional database. For this study, we considered data of patients with histological and clinical diagnosis of CC and treated with NACT + RS at our Unit of Gynecologic Oncology, Policlinico Umberto I, Sapienza University of Rome. Patients have been treated over a 10-year period, between January 2012 and December 2022. After revision of all records, patients were restaged according to FIGO staging system 2018 [15].

Inclusion criteria were: 1) age >18 and <75 years; 2) histological diagnosis of cervical squamous cell carcinoma or adenocarcinoma; 3) presence of positive lymph nodes at imaging (FIGO IIIC1-2 stage); 4) no previous oncological treatments; 5) no actual pregnancy; 6) World Health Organization performance status 0–2 [16]; life expectancy of at least 3 months; written informed consent to research project; no evidence of distant metastasis or concomitant neoplasms.

All clinical, histopathological and imaging data were collected in a dedicated database.

Staging procedures included [3,15]: clinical examination performed by an expert gynecologic oncologist under anesthesia with bimanual exam (vaginal and rectal) and cervical biopsies/conization for histological assessment; in case of suspected involvement of the rectum or bladder cystoscopy and/or proctoscopy were performed; blood chemistry tests and CA125; chest computed tomography (CT) scan, pelvic transabdominal/2D/3D and transvaginal ultrasound with Power Doppler [17], abdomino-pelvic magnetic resonance imaging (MRI) and/or positron emission tomography (PET)-CT to assess local spread (parametria, vaginal, rectal, bladder), nodal involvement (pelvic and/or aortic) and distant metastases. Diagnostic laparoscopy with peritoneal biopsies was proposed in selected cases of positive aortic node or massive involvement of anterior/posterior parametria or suspicious peritoneal involvement to exclude peritoneal involvement [18].

Patients were submitted to 3 cycles of platinum-based NACT (cisplatin 75 mg/m² or carboplatin area under the curve=5 plus paclitaxel 175 mg/m²) every 21 days (or weekly regimen in selected cases, mainly for frail patients), with a first clinical evaluation of response was performed after 2 cycles to early identify non responder diseases. After NACT all patients underwent abdomino-pelvic MRI and pelvic transabdominal and transvaginal/2D/3D transvaginal-ultrasound (US) with Power Doppler to evaluate response to chemotherapy [19]. Patients with progression of disease after NACT were excluded and sent to CCRT-B.

Response to NACT was assessed following the Response Evaluation Criteria in Solid Tumors (RECIST) criteria version 1.1 for solid tumors [20]. All patients who were considered eligible for surgery after NACT (complete response with no residual tumor or partial response according to RECIST criteria) underwent radical hysterectomy (type B, C or D, depending on nodal status) [21] with bilateral salpingo-oophorectomy and systematic pelvic lymphadenectomy with curative intent. Aortic lymphadenectomy was performed in case of imaging positive for aortic disease, intraoperative bulky nodal disease or grossly metastatic pelvic nodal disease. If necessary, additional surgical procedures were conducted, including ureteral resection, bladder resection, uretero-ureteral anastomosis, and ureteral reimplantation. Adjuvant therapy was administered in case of positive parametrial or vaginal surgical margins (radiotherapy [RT]/brachytherapy) and in case of positive lymph nodes (chemotherapy alone or chemoradiation in case of ≥3 positive lymph nodes). A patient could have one or more risk factors to consider for choosing the correct adjuvant treatment.

At the end of the treatment, patients started a follow-up program every 3 months for the first 2 years, then every 6 months until the fifth year. Clinical examination and colposcopy were conducted at each visit. Pap smear test every 6 months for the first 2 years. Moreover, patients in follow-up were evaluated by a total body CT scan every 6 months for the first 2 years than alternately by abdominal US until the fifth year. After the fifth year, patients were evaluated annually with a clinical examination, colposcopy and ultrasound, and more advanced imaging tests (thoracic/chest CT, PET-CT, MRI) were requested only in case of symptoms. Patterns of recurrence were recorded. Recurrent disease was treated by surgery, RT or chemotherapy according to the specific indication. Pattern of recurrence and its treatment was detailed for each patient. Data on overall survival and progression free survival were calculated.

3. Statistical analysis

Descriptive statistics were used to summarize the baseline characteristics of the study population. Continuous variables were reported as medians with interquartile ranges (IQR), while categorical variables were summarized as frequencies and percentages. Survival analyses were conducted to assess overall survival (OS), progression-free survival (PFS), and OS after first relapse. Kaplan-Meier curves were generated to estimate survival probabilities for each outcome. The log-rank test was used to compare survival distributions across groups, and p-values <0.05 were considered statistically significant. Results were censored at the last follow-up for patients who had not experienced the event of interest. Median survival times and 95% confidence intervals (CIs) were reported where applicable. All analyses were performed using R (version 4.4.1; R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

From January 2012 to December 2022, 339 patients affected by CC were treated at the Department of Maternal and Child Health and Urological Sciences, Policlinico Umberto I, Sapienza University of Rome. After evaluation of inclusion criteria data from 97 patients were collected (Fig. 1).

Fig. 1. Selection of the sample flowchart.

Patients’ characteristics are described in Table 1. Briefly, the prevalent histotype was squamous cell carcinoma, 84/97 patients (86.6%), and 88/97 patients (90.7%) were FIGO stage IIIC1. After NACT, 3 patients (3.1%) showed a complete radiological response on cervical disease, 86 patients (88.7%) were classified as partial responders and 8 (8.2%) patients had stable disease according to RECIST criteria version 1.1 [20]. No patients experienced progressive disease.

Table 1. Characteristics of the patients.

Characteristics		Values
Patients		97 (100.0)
Age at the diagnosis (yr)		52 (31–75)
BMI (kg/m2)		24 (14–40)
Histology
	Squamous cell carcinoma	84 (86.6)
	Adenocarcinoma	11 (11.3)
	Neuroendocrine	2 (2.1)
FIGO stage
	IIIC1	88 (90.7)
	IIIC2	9 (9.3)
Grading
	G1	5 (5.2)
	G2	33 (34.0)
	G3	59 (60.8)
Diameter of the lesion (mm)		50.5 (20–96)
Diameter of largest involved pelvic node (MRI based in mm)		18 (8–42)
Diameter of largest involved aortic node (MRI based in mm)		15 (8–25)

Values are presented as number (%) or median (range).

BMI, body index mass; FIGO, International Federation of Gynecology and Obstetrics; MRI, magnetic resonance imaging.

Thirty-four patients (35.1%) showed complete disappearance of nodal disease, and 49 patients (50.5%) had a reduction in size of nodal metastasis (Table 2). The primary treatment was type C radical hysterectomy (n=59, 60.8%) with systematic pelvic lymphadenectomy (Fig. 2); in 17 cases (17.5%) paraaortic lymphadenectomy was performed. Additional surgical procedures were necessary in 16 patients (16.5%). Among them, 11 required ureteral resection, with 7 undergoing uretero-ureteral anastomosis and 4 having ureteral reimplantation. Partial full-thickness bladder resection was performed in 2 patients, while 3 underwent extramucosal bladder resection. Intraoperative complications included an estimated blood loss of over 1,000 mL in 10 patients (10.3%), with 6 patients (6.2%) requiring intraoperative blood transfusions. In the early postoperative period, anemia (hemoglobin <8 g/dL) was observed in 8 patients (8.2%), while 11 patients (11%) developed a postoperative fever. Wound infections occurred in 7 cases (7.2%), and urinary tract fistulas (G3) were noted in 5 patients (5.2%) approximately 10 days after surgery [22]. The median hospital stay was 6 days, ranging from 4 to 12 days.

Table 2. Radiological response to NACT.

Variables		Values
Tumor (RECIST criteria)
	Complete response	3 (3.1)
	Partial response	86 (88.7)
	Steady disease	8 (8.2)
	Progression disease	-
Nodal
	Complete disappearance of nodal metastasis	34 (35.1)
	Reduction in size of the nodal metastasis	49 (50.5)
	Stable/progressive disease	14 (14.4)

Values are presented as number (%).

NACT, neoadjuvant chemotherapy; RECIST, Response Evaluation Criteria in Solid Tumors.

Fig. 2. Radical hysterectomy and pelvic lymphadenectomy. (A) Surgical specimen after radical hysterectomy. (B) Surgical view after systematic pelvic lymphadenectomy.

Complete surgical characteristics and pathological data are shown in Table 3.

Table 3. Surgical characteristics and pathological data after surgery.

Surgical characteristics		Values
Radical hysterectomy*
	Type B	22 (23.7)
	Type C	59 (60.8)
	Type D	16 (16.5)
Lymphadenectomy
	Pelvic	97 (100.0)
	Paraaortic	17 (17.5)
Histological features
	Median tumor diameter (mm, range)	29 (0–65)
	Positive vaginal or parametrial margins	10 (10.3)
	Negative nodes	54 (55.7)
	Positive lymph nodes	43 (44.3)
	Pelvic	34 (35.1)
	Pelvic + paraaortic	9 (9.3)
	Parametrial involvement	22 (22.7)

Values are presented as number (%) or median (range).

^*According to Querlew-Morrow Classification.

Regarding adjuvant treatment and recurrence, after surgery 27 (27.8%) patients underwent adjuvant chemotherapy alone, 17 patients (17.5%) received adjuvant chemoradiotherapy, 8 patients (8.2%) were submitted to RT; 45 patients (46.4%) did not receive any adjuvant treatment. In total 46 patients (47.4%) experienced recurrence: 9 patients (9.3%) at a single site and 37 patients (38.1%) at multiple locations. The most frequent sites of recurrence were the pelvic nodes (n=21, 21.6%) and the central pelvis (n=9, 9.3%). Eleven patients (23.9%) experienced recurrence as distant metastasis. Five patients (10,8%) had vaginal recurrence. No patients experienced lateral pelvic recurrence.

Treatment of recurrence was as follows. Among the 46 patients with recurrence, 21 patients (45.7%) received palliative chemotherapy while 25 patients (54.3%) were treated with a curative intent. Specifically, 10 patients (21.7%) were submitted to surgery, 7 patients (15.2%) underwent chemoradiotherapy, 7 (15.2%) patients received a combined treatment with CT plus surgery, and 1 patient (2.2%) underwent RT. After a median follow up of 51 months (range, 11–136 months), 32 (33.0%) patients are dead of disease, 5 (5.2%) are dead of other diseases, 56 (57.7%) are alive without disease and 4 (4.1%) are alive with disease and under treatment.

Survival data are shown in Fig. 3 (OS and PFS data). Specifically, 2-year OS of the all population was 0.801 (95% CI=0.725–0.885); 5-year OS was 0.611 (95% CI=0.514–0.725); 2-year PFS was 0.695 (95% CI=0.607–0.797) while 5-year PFS was 0.585 (95% CI=0.487–0.702). Survival data have been expressed also in correlation with the FIGO stage (IIIC1 and IIIC2); in particular, 2-year OS for stage IIIC1 was 0.819 (95% CI=0.742–0.903); 5-year OS for stage IIIC1 was 0.614 (95% CI=0.513–0.734). On the other hand, 2-year OS for stage IIIC2 was 0.571 (95% CI=0.301–1.000), while 5-year OS for stage IIIC2 was 0.571 (95% CI=0.301–1.000) (the curve remains constant from 2 to 5 years) with hazard ratio (HR) for OS in stage IIIC2 of 1.280 (95% CI=0.391–4.189), with a p-value of 0.683. Moreover, 2-year PFS for stage IIIC1 was 0.718 (95% CI=0.628–0.821) and 5-year PFS for stage IIIC1 was 0.597 (95% CI=0.494–0.720). While for stage IIIC2 the PFS at 2-year was 0.429 (95% CI=0.182–1.000), while at 5-year was 0.429 (95% CI=0.182–1.000) (the curve remains constant from 2 to 5 years) with an HR for PFS in stage IIIC2 of 1.769 (95% CI=0.624–5.005), with a p-value of 0.283. Moreover, the median OS after first recurrence was 1.01 years, considering 1-year OS after first recurrence 0.531 (95% CI=0.403–0.701) while 2-year OS after first recurrence 0.362 (95% CI=0.243–0.539). The study also performed a survival analysis based on the sites of recurrence. Among patients with vaginal recurrence, no deaths were recorded. The 2-year OS rate for those with lymph node recurrence was 0.700 (95% CI=0.525–0.933). In contrast, the 2-year OS for central pelvic recurrence and metastatic recurrence were 0.455 (95% CI=0.238–0.868) and 0.416 (95% CI=0.199–0.868), respectively, with a p-value of <0.001.

Fig. 3. Survival data. (A) PFS of all population. (B) OS of all population.

OS, overall survival; PFS, progression-free survival.

DISCUSSION

The idea of a study focusing on patients affected by high risk LACC (i.e., positive nodes patients, FIGO IIIC1–2) at diagnosis arose from some considerations. First, nodal metastasis has been recognized as the most powerful prognostic factor in CC patients [23] and, consequently, stages IIIC1 and IIIC2 have been added to the 2018 FIGO staging classification (nodal involvement detected either clinically or pathologically) [15].

Secondly, survival of stage IIIC1 and IIIC2 patients, treated with standard CCRT-B, is reported to be 60.8% and 37.5% respectively in a large national database [24]; however, it is much more heterogeneous also influenced by tumor volume, with 45% and 23% OS in patients with tumor diameter ≥4 cm respectively in IIIC1 and IIIC2 stage [25].

Third, the quality of life of CC patients treated with CCRT-B is poor [26], with vaginal stenosis representing the most disabling and irreversible condition [27], even if in the last two decades the advances in techniques and equipment including the 3D-image-guided brachytherapy reduced the risk of acute and late morbidity and improve survival data [6]; however, this specific expertise and tools are not available in all the centers, and even more in developing countries, where CC is more common.

Fourth, before the introduction of antiangiogenic and immunotherapy, a relapse after CCRT-B represented a fatal event, with a poor median survival, 7–10 months in some published experience [14], substantially because a salvage surgery was mostly not feasible or, more frequently, for the occurrence of distant metastasis [12,13].

Last, evidence coming from past trials suggest that CC is a chemosensitive tumor [28]. Many studies have shown that NACT before radical surgery can shrink the tumor size, control distant metastasis, reduce positive nodes, parametrial infiltration and, in particular, post operative RT compared with surgery alone [29]. In the present study, at the end of NACT, median cervical tumor size was significantly reduced (pre-NACT 50 mm, range 20–96 mm vs. post-NACT 28 mm, range 0–65 mm) and radiologically suspected nodal metastasis reduced in size or disappeared respectively in 51% and 35% of cases. Even though multiple studies and meta-analyses have explored NACT in LACC, the most appropriate timing and sequence of therapy remain debated. Whether selected patients might derive greater benefit from immediate radical treatment or from surgery after NACT continues to be an open question [30,31].

Indeed, in advanced stages of CC, an individualized treatment strategy may be preferable to a uniform approach. Patients presenting with isolated nodal involvement and limited pelvic disease could potentially benefit from NACT followed by surgery, whereas those with more extensive parametrial or pelvic wall invasion might be better suited for concurrent chemoradiation. These considerations suggest that treatment selection should be guided by disease extent, biological behavior, and institutional resources, aiming to avoid overtreatment and to preserve functional outcomes whenever feasible.

Recently, some authors showed that either induction chemotherapy [32] and the addition of immunotherapy to standard chemoradiation treatment [33] improve survival in LACC patients. Moreover, neoadjuvant chemo-immunotherapy combining cisplatin, nab-paclitaxel, and camrelizumab was evaluated in a recently published single-arm, phase 2 trial (NACI study) in JAMA Oncology. The trial enrolled 85 patients with programmed death-ligand 1-positive LACC (FIGO stages IB3, IIA2, IIB/IIICr) and demonstrated a high objective response rate of 98%, with 81 patients ultimately undergoing radical surgery [34]. Although immunotherapy in the neoadjuvant setting is not currently part of standard clinical practice, these early data open promising perspectives toward a more personalized approach to the management of LACC and may offer a promising alternative to standard chemoradiotherapy in selected patients.

In this study, the 5-year OS and PFS rates for node-positive CC patients treated with NACT followed by RS were 61% and 59% for stage IIIC1, and 57% and 43% for stage IIIC2, respectively, comparable to those reported in the literature for standard CCRT-B. However the pattern of relapse was different, and we were able to propose in 25 out of 46 (54.3%) recurrent patients a curative procedure either surgery, radiation alone or chemoradiation.

Although postoperative morbidity was not a primary endpoint of this study, several reports have investigated the relationship between surgical radicality and postoperative functional outcomes after NACT and radical hysterectomy. These studies consistently indicate that extended (type C2–D) resections are associated with higher rates of long-term bladder and sexual dysfunction compared with nerve-sparing approaches, highlighting the need for an individualized balance between radicality and morbidity when surgery follows NACT [31,35,36].

This study has several limits: the retrospective design, the lack of nodal pathological assessment before NACT, the unavailability of antiangiogenic/immunotherapy at that time, the lack of recorded data regarding long term complications and quality of life, the scarce reproducibility of such an approach, which require a very high expertise in gynecological pelvic surgery.

On the other hand, in our opinion, in this specific era of precision medicine, and with the availability of immunotherapy, we can hypothesize to identify patients who can better respond to systemic therapy [37] in order to reduce the tumor burden before definitive local treatment, which can be either a good RT or a good surgery depending on the resources and expertise of the center.

Some considerations should be reserved also to the current recommendation of adjuvant treatment, still based on Sedlis et al. [38] and Peters et al. [39]’s criteria which dated back in 1999–2000, and applied also to LACC without the support of randomized controlled trials; for sure the combo surgery plus RT should be avoided as much as possible in CC patients in order to avoid treatment related long term complications as described in literature [27]; indeed, in our series we propose RT only in case of positive margins (vaginal or parametrial) or multiple nodal metastasis, in order to improve local control. Some studies have been published on adjuvant chemotherapy alone with encouraging results [40].

In conclusion, since the standard treatment offers about 60% chance of survival for patients in the high-risk category (FIGO IIIC1–2), along with potentially poor quality of life, especially for younger women, it is reasonable to explore alternative approaches. The present findings indicate that NACT + RS may serve as a feasible alternative to chemoradiotherapy for patients with LACC and lymph node involvement (FIGO stage IIIC1–2), showing 5-year OS and PFS rates that are comparable to those reported for standard treatment. Additionally, recurrence patterns, mainly localized to pelvic lymph nodes or the central pelvis, allowed for curative interventions in over 50% of cases. The surgical approach in LACC represents an evolution of modern pelvic surgery, as radical hysterectomy remains one of the most complex procedures in gynecologic oncology. Adequate training and high-level expertise are essential, since several studies have confirmed that outcomes are significantly influenced by surgical skill and institutional experience, both in terms of survival and postoperative quality of life. Although standardization in such complex surgery is inherently challenging, this should not preclude the possibility of offering highly individualized and technically optimal procedures. At the same time, while concurrent chemoradiation is, by its nature, easier to standardize and remains the mainstay of treatment in many settings, its potential long-term impact on quality of life, particularly in younger women, where late toxicity and functional sequelae may be clinically meaningful, should not be overlooked. Future integration with novel systemic therapies, such as immunotherapy, may further enhance outcomes, presenting new possibilities in the personalized treatment of advanced CC. It may not yet be time to abandon neoadjuvant therapy combined with surgery as a treatment option for LACC.