Fertility Preservation Approaches

The average age of patients with newly diagnosed invasive cervical cancer is approximately 50 years; thus, a significant portion of patients are of childbearing age. For these patients with IA1, IA2, IB1, and select IB2 disease (less than 2 cm in maximal diameter), fertility-sparing approaches may be considered for those who wish to preserve childbearing. The designation of FIGO stage IA1 indicates early-stage, microinvasive disease ≤3 mm in depth of invasion, and without evidence of lymphovascular space invasion (LVSI). These patients may be managed conservatively with cervical diagnostic excisional procedures such as a cone biopsy or conization, or alternatively using a loop electrosurgical excision procedure (LEEP). These procedures are both diagnostic and therapeutic, seeking to achieve complete en bloc removal of the suspicious lesion for adequate pathological evaluation of depth, LVSI, margin status, HPV staining, and histological subtyping.

LEEP and Conization

In cold knife conization, a scalpel is used to resect the specimen. Gynecologists may use acetic acid solution or Lugol’s iodine to better visualize the borders of the lesion. Vasoconstrictor solutions such as vasopressin may be injected into the cervix to reduce intra-operative bleeding. An exocervical circumferential incision is taken with the sharp blade completely around the lesion and ectocervix, then deeper in a conical shape until the lesion and endocervical canal are removed en bloc by severing the specimen at the apex of the cone. The mucosal surfaces of the ectocervix, transformation zone, and lesion are not disturbed. Of note, the apex of the cone may be taken deeper in older patients as the transformation zone often moves cranially with advancing age. While endocervical curettage is routinely performed after conization, formal dilation and curettage is not performed unless there is suspicion for a uterine pathology. Management of the cone bed varies, and ranges from gauze packing, tampon application, electrocautery, sutures, or other hemostatic agents. LEEP uses modern electrosurgical current conducted through a wired loop to cut through tissue instead of a scalpel. Prior studies suggest no obvious inferiority of one technique over the other (LEEP or cone biopsy) for eradication of disease or for bleeding.5 One distinct advantage of LEEP is its potential to be used in the outpatient clinic while cone biopsies primarily take place in procedure suites under anesthesia. LEEP generators can also be tuned or “blended” to balance cutting efficacy and coagulative effect of the electrical current. However, if LEEP is utilized, care should be taken to allow for intact evaluation of specimen margins with minimal electrosurgical artifact (thermal damage). Non-conductive speculums must be used to avoid damaging the vaginal canal. In contrast to scalpel-based conization which uses a circumferential incision, LEEP uses a “scooping” motion to carve around and under the transformation zone and the lesion. Occasionally, the apex of the cone is surgically cut to minimize electrocoagulation artifact. The surgeon should not fragment the specimen if possible. The surgeon may choose from a variety of different loop tips with different shapes and diameters to obtain the specimen. In this way, like conization, the gynecologist is able to use judgment to take a wider resection or a narrower, deeper cone resection based on the shape and location of the lesion. Another pertinent consideration for the surgeon is the propensity of adenocarcinomas to have “skip” lesions or multifocal disease involving the deep endocervical canal or deeper portions of endocervical clefts in up to 10%–15% of specimens.6 7 Historically, ablative laser therapy was also used to completely eradicate small superficial lesions; however, this practice is no longer recommended as it does not produce an evaluable specimen for the pathologist. A laser-based conization procedure has also been used by specialists but is a more involved procedure and requires additional expertize.8 9

In FIGO IA1 specimens without LVSI, risk of nodal disease is minimal (<1%) and typically no sentinel lymph node (SLN) mapping or pelvic lymphadenectomy is indicated.10–12 For patients past reproductive age or not interested in fertility preservation and medically fit, extrafascial/simple hysterectomy (Type A) may also be offered as definitive treatment via a vaginal, laparoscopic, or minimally invasive approach. For specimens with microscopic or microinvasive disease (FIGO IA1), pelvic examination and annual cervical cytology are the primary means of surveillance.

The oncologic outcomes of conservative procedures such as conization plus surgical lymph node assessment for stage IA2 and IB1 tumors are a current area of investigation. The multicenter single-arm prospective ConCerv trial recently published their results after accrual of 100 patients with grade 1 to 2 stage IA2 to IB1 tumors with less than 10 mm depth of invasion, without LVSI, and with negative margins after conization, allowing patients to choose fertility preservation (surgical nodal assessment only) or simple hysterectomy plus nodal assessment. The authors reported a lymph node involvement rate of 5% in this cohort (of which 67% were IB1 tumors) and a 2-year failure rate of 3.5%.13

Radical Trachelectomy

In addition to IA1 patients (discussed above), patients with IA2 and IB1 disease are also typically eligible for fertility preservation surgery if desired, namely radical trachelectomy. Simple trachelectomy removes the majority of the cervix but does not take dissection beyond the anatomical borders of the cervix itself and thus should be reserved only for those with in situ disease or IA1 disease. For those with IA2 and IB1 disease and tumors of 2 cm or less, a radical trachelectomy includes resection of most of the cervix as well as additional oncologic margin. Specifically, most of the cervix is removed, and the dissection should be taken an additional 1–2 cm along the vaginal fornices with parametrial resection, the ureters visualized and dissected from the cervix, and uterosacral ligaments divided 1–2 cm dorsal to the cervix, preserving the hypogastric nerve plexus. The bladder is mobilized to the upper vagina and the rectum is mobilized below the cervix. A small (approximately 5 mm) remnant of the cervix may be left for anchoring of the cerclage during future pregnancy. Compared with microinvasive disease, IA2 and IB1 patients are at increasing risk of lymphatic metastases, and pelvic lymphadenectomy with SLN mapping should be performed accompanying the radical trachelectomy, and is discussed separately.14–16

Open abdominal approach is regarded as a standard of care for radical trachelectomy, though vaginal approaches can be used for select patients with tumors less than 2 cm (up to IB1).17 18 The main advantage of an open abdominal approach is improved access to the parametria.17 19 Institutional series have reported outcomes of radical trachelectomy in patients with tumors between 2 cm and 4 cm with mixed results. A case series of 29 patients with tumors between 2 cm and 4 cm underwent trachelectomy and after a median follow-up of 44 months only one had relapse; however, 45% underwent completion hysterectomy due to additional risk factors and 21% went on to receive adjuvant chemoradiation.20 One study identified that early-stage patients with tumors larger than 2 cm had significantly worse 5-year disease-free survival after trachelectomy compared with those with tumors less than 2 cm, an absolute difference of 13% (p=0.039).21 These studies underline the increased recurrence risk of tumors between 2 cm and 4 cm for trachelectomy and, as such, hysterectomy with pelvic nodal dissection remains standard of care. Recommendations for trachelectomy in these patients remain guarded.

More recently, institutions have implemented minimally invasive approaches (laparoscopic and robotic) to trachelectomy though these results have not been directly compared with open trachelectomy. An ongoing multi-institutional retrospective analysis led by MD Anderson seeks to compare outcomes between open and minimally invasive approaches (laparoscopic or robotic) to radical trachelectomy.22 During patient education and informed consent, patients should be made aware that pregnancy following trachelectomy portends an increased risk of miscarriage and pre-term labor. Successful delivery rates are near ~60%.23–26

Regarding non-squamous, non-adenocarcinoma, and non-adenosquamous histologies: cervical histologies such as small cell/neuroendocrine tumors, gastric-type adenocarcinoma, or adenoma malignum are not suitable for fertility-sparing surgical approaches, even radical trachelectomy, given their particularly aggressive nature and lack of evidence demonstrating efficacy.27 For surveillance after fertility-preservation treatment of early-stage disease, interval evaluation is recommended 3–6 months for 2 years and then every 6–12 months up to 5 years then annually by the National Comprehensive Cancer Network (NCCN). Cervical cytology should be obtained at least annually (Table 2) though its added value following radical trachelectomy is unclear to date.28

Definitive chemoradiation may be recommended for cervical cancer patients whose stage indicates a hysterectomy but are not medically fit for the procedure. Alternatively, radiotherapy alone may be considered for early-stage but medically inoperable patients.

Non-Fertility-Sparing Approaches

For stage IA2 to IB1 patients, past childbearing age or not interested in fertility-sparing, or those with tumors greater than 2 cm (IB2), a modified radical (Type B) or radical (Type C1) hysterectomy with pelvic lymphadenectomy is the standard of care. The surgeon’s choice of a modified radical versus radical hysterectomy is based on the extent of the lesion. For instance, for stage IA1 with LVSI or IA2 disease a modified radical approach (Querleu/Morrow Class B) may be most appropriate, while for IB1/2 disease or larger a radical hysterectomy may be required (Querleu/Morrow Class C1/C2). In these cases, the intent of hysterectomy is always curative with the goal to remove the entire extent of disease with negative margin, the entire cervix, and uterus. Simultaneous oophorectomies are optional and may depend on various factors such as age and menopausal status. Some centers consider adenocarcinoma histology indication for simultaneous oophorectomies, though this is controversial. Detailed descriptions of the procedures are defined in the Querleu and Morrow surgical classification of 200829 which has supplanted the Piver-Rutledge-Smith classifications and are briefly summarized below.29

As discussed above, a modified radical hysterectomy completely removes the uterus and cervix via an open laparotomy incision. A vaginal margin of 1–2 cm is recommended, and the bilateral ureters are unroofed and dissected away from the cervix, and the parametria are resected also by 1–2 cm horizontally, as are the uterosacral ligaments. The bladder is mobilized to the upper vagina and the rectum is mobilized below the removed cervix. The radical (Type C1) hysterectomy is a more extensive procedure and requires more adjacent dissection. Key differences include a larger vaginal margin resection, removing up to the upper one-third of the vagina, and the bladder mobilized to the mid-vagina and the rectum mobilized below the mid-vagina. The entire parametria is dissected out to the internal iliac vessels which is particularly crucial for tumors with suspected lateral involvement of the parametria. The C1 radical hysterectomy also seeks to preserve the hypogastric nerve plexus (nerve sparing) and divides the uterosacral ligaments at least 2 cm dorsal to the cervix. Select IIA1 disease may also be considered for hysterectomy, depending on resectability of the vaginal component of disease and tumor size (<4 cm). Those with IB3 (>4 cm) or greater disease should be referred to definitive chemoradiotherapy, though some foreign practices patterns still consider upfront surgery standard for IB3 tumors.

Abdominal Radical Hysterectomy Versus Minimally Invasive Surgery

In more recent years, greater emphasis has been placed on the recommendation of abdominal radical hysterectomy versus minimally invasive/laparoscopic approaches. In a prospective randomized clinical trial sponsored by Medtronic and MD Anderson with the goal of evaluating for noninferiority, 631 patients were randomly assigned to either open surgery or minimally invasive laparoscopic approach to radical hysterectomy. The arms were well balanced with respect to FIGO stage, histology, age, and performance status. The majority (92%) of patients were staged as “old” IB1 based on the 2009 FIGO staging (tumors <4 cm), which would now encompass both IB1 and IB2, and exclude IB3 (tumors >4 cm) in the 2018 FIGO staging. Disease-free survival evaluated at 4.5 years (the primary endpoint) revealed a significant 10.6% absolute decrease in disease-free survival (86% vs 96.5%) in the minimally invasive approach. Recurrences were most common in the vaginal vault or the pelvis. Interestingly, non-vault pelvic recurrences only occurred in the minimally invasive group. Additionally, the authors observed an absolute decrease of 5.2% in 3-year overall survival (93.8% vs 99%) and an absolute increase of 3.8% in the 3-year rate of death from cervical cancer (4.4% vs 0.6%). The authors concluded that minimally invasive approaches to radical hysterectomy in this patient population were associated with a detriment to disease-specific survival and overall survival.30 Complementary to this finding is a systematic review and meta-analysis performed of over 9000 patients at 15 centers which yielded a HR of 1.71 (71% higher likelihood of recurrence or death) in those who underwent minimally invasive hysterectomy compared with those who underwent open surgery.31 As a counterpoint to these studies, the aforementioned ConCerv trial authors noted that their favorable failure rate of 3.5% (although with shorter follow-up) was achieved in a cohort wherein 96% of the patients received a minimally invasive surgical approach. There is an ongoing prospective trial (SHAPE, NCT01658930) which is investigating the potential role of simple hysterectomy in low-risk patients versus standard radical hysterectomy; however, this study currently does not stratify patients according to minimally invasive versus open approach.

Pelvic Lymph Node Assessment

In patients with stage IA1 disease only without additional risk factors, fertility-sparing or non-fertility sparing approach to surgery is sufficient. However, for stage IA1 with LVSI, or any stage IA2 or higher, surgical pelvic lymph nodal assessment is indicated as even these early-stage patients may harbor occult nodal disease in greater than 5% of cases and typically consists of SLN mapping and pelvic lymphadenectomy.32 To date there have been no high-quality prospective data supporting the omission of pelvic lymph node sampling in these patients. Surgical techniques for nodal sampling are not discussed in detail in this review.

Post-Operative Pelvic Irradiation

After undergoing modified radical hysterectomy, surgical specimens are assessed for risk factors that may warrant adjuvant therapy. GOG 92 was a phase III randomized clinical trial comparing no further treatment versus adjuvant radiation following radical hysterectomy and lymph node dissection in patients with stage IB, node-negative disease with certain risk factors. Eligible patients were: (1) LVSI present, deep one-third stromal invasion, with any tumor size; (2) LVSI present, middle one-third stromal invasion, tumor size ≥2 cm; (3) LVSI present, superficial one-third stromal invasion, tumor size ≥5 cm; or (4) LVSI absent, with deep or middle one-third stromal invasion, tumor size ≥4 cm. The study accrued 277 evaluable patients randomized in a 1:1 fashion and demonstrated superior oncologic outcomes with adjuvant radiotherapy. Ten-year local recurrence rates were reduced from 21% to 14% with the addition of radiotherapy and 10-year overall survival trended in favor of radiotherapy by 9% (71% vs 80%, p=0.074).33 To date, the current standard practice is to offer post-operative pelvic radiotherapy in this patient population.

Post-Operative Chemoradiation

There exist several randomized studies to inform on which patients may benefit from the addition of chemotherapy to adjuvant radiotherapy. Though study populations remain rather heterogenous, some conclusions may be drawn regarding optimal patient selection. The GOG 109/SWOG 8797/RTOG 9112 study was a phase III clinical trial that assessed the role of adjuvant chemoradiation versus radiation alone in high-risk early-stage cervical cancer. High-risk was defined as either stage IA1, IB, IIA post-radical hysterectomy and pelvic lymph node dissection, with either positive pelvic lymph nodes, parametrial involvement, or positive surgical margins. Patients in the intervention arm received cisplatin 70 mg/m² + 5-flououracil 1000 mg/m², every 3 weeks for four cycles, (two cycles concurrently and two adjuvantly). The addition of four cycles of cisplatin and 5-flououracil yielded a 4-year progression-free survival of 80% compared with 63% for radiotherapy alone, and 4-year overall survival benefit of 10% (81% compared with 71% with radiotherapy alone).34

The STARS trial was a Chinese phase III clinical trial of over 1000 patients first published in 2021 that employed a somewhat broader inclusion criteria but also asking the question of adjuvant radiotherapy alone versus chemoradiation (weekly cisplatin), and versus a third arm of sequential “sandwich” chemotherapy (two cycles of cisplatin and paclitaxel, followed by radiotherapy, followed by two additional chemotherapy cycles). Patients on study included those receiving hysterectomy and pelvic lymph node dissection with stage IB1 to IIA2 disease and any one or more of the following risk factors: LVSI, deep stromal invasion, pelvic node-positive, positive margins, or parametrial invasion (92.3% had open surgical approach). This study showed improved disease-free survival and overall survival at 3 years with sequential chemoradiation compared with radiation alone (disease-free survival 90% vs 82% with radiotherapy alone, p=0.01), and improved disease-free survival compared with concurrent chemoradiotherapy (90% vs 85% with chemoradiotherapy, p=0.04).35 At a median follow-up of 4.7 years, distant metastases were nearly halved on the sandwich regimen versus the other two groups (6.5% sandwich, 10.6% radiotherapy, 11% chemoradiotherapy; p=0.05 and 0.04, respectively). The sandwich regimen yielded a modest but significant 5-year overall survival benefit over radiotherapy alone of 4% and 5% in the intention-to-treat analysis and per-protocol analysis, respectively. Sandwich therapy trended towards improved survival versus chemoradiotherapy on per-protocol analysis but was non-significant (p=0.07). Because STARS utilized a broader inclusion criterion, some patients suitable for GOG 92 may have also been eligible for STARS, thus blurring the early-stage indications for chemotherapy. The ongoing GOG 0263 trial seeks to refine patient selection for concurrent systemic therapy by assessing the role of adjuvant chemoradiation versus radiation alone in patients with stage I to II disease and intermediate-risk factors (deep one-third invasion; middle one-third and tumor ≥2 cm; or superficial one-third and tumor ≥5 cm).36

In contrast to STARS, the OUTBACK trial (NCT01414608) evaluated additional chemotherapy starting after chemoradiotherapy, and enrolled 919 patients. Initial 5-year data have been presented in abstract form. The study included both early and locally advanced cervical cancer (FIGO 2008 stage IB1 and node-positive, IB2, II, IIIB, or IVA) and showed preliminarily that there was no benefit in progression-free survival with the addition of outback chemotherapy (63% outback vs 61% chemoradiotherapy), and 5-year overall survival were similar (72% outback vs 71% chemoradiotherapy). These results may have tentatively curbed some institutional practices of additional adjuvant chemotherapy after chemoradiotherapy, though the results of STARS show some promise of early initiation of chemotherapy at reducing distant failures. The ongoing RTOG 0724 trial is assessing the role of outback chemotherapy (paclitaxel and carboplatin) in patients receiving concurrent chemoradiation with weekly cisplatin in patients with cervical cancer stage IA2, IB, and IIA post-radical hysterectomy with either positive pelvic nodes, positive parametrial involvement, or positive para-aortic lymph nodes (PET negative after surgery).37 The current clear indications for post-operative chemoradiation in patients include positive pelvic lymph nodes, parametrial involvement, and positive surgical margins unfit for re-resection, and a summary of indications for adjuvant treatment following surgery is given in Box 1. Of note, there is ongoing investigation (INTERLACE trial) regarding the role of neoadjuvant induction chemotherapy using carboplatin and paclitaxel in a predominantly locally-advanced population, but this does allow patients with stage IB2 to IVa disease.38

Intensity Modulated Radiation Therapy (IMRT) Reduces Acute and Chronic Toxicity

Two-dimensional (2D) or three-dimensional conformal radiation (3DCRT) has been the historical standard radiation treatment modality for post-operative management of cervical cancer for decades and much of the early 2000s. Potential chronic toxicities of pelvic irradiation include diarrhea, urinary urgency or frequency, hematuria, as well as potential urinary or bowel incontinence. The implementation of modern intensity modulation techniques (IMRT) lowers high-dose radiation to adjacent organs at risk and has led to significant benefits in toxicity profile. RTOG 1203 (TIME-C) was a phase III trial that compared 3DCRT to IMRT in women with cervical or endometrial cancer requiring post-operative pelvic radiation or chemoradiation with a primary endpoint of acute gastrointestinal toxicity. IMRT showed fewer acute grade 2+ gastrointestinal toxicity with overall lower incidences of diarrhea, fecal incontinence, and need for anti-diarrheal medications. Long-term results from the TIME-C trial showed reductions in chronic abdominal pain, diarrhea, and fecal incontinence in the IMRT arm.39 The PARCER trial was another phase III trial that compared 3DCRT versus IMRT in cervical cancer patients with a primary endpoint of late gastrointestinal toxicity. This study also showed reduced incidence of acute diarrhea during treatment in the IMRT arm and reduced late gastrointestinal toxicity such as anorexia, abdominal bloating, and bowel obstruction. The TIME-C and PARCER trials have jointly established the role of IMRT as the standard treatment option for post-operative pelvic radiation in gynecologic malignancies.40 On a technical note, in the instance that there is significant motion of the target and/or adjacent organs, some institutions have employed an adaptive “library of plans” technique in which a variety of plans are generated based on observed or predicted variability in target motion and bladder filling, and before each fraction a “plan-of-the-day” is selected which best fits the current patient setup, thus reducing total irradiated volume and cumulative dose to organs at risk.41 42

Brachytherapy

In tandem with reductions in toxicity after external beam radiation therapy using IMRT, advances in 3D image-based brachytherapy have also led to further reductions in toxicity and even boasts an improvement in local control. The French “STIC” study found that by using a 3D planning methodology the authors were able to reduce grade 3+ toxicities by nearly half when compared with conventional 2D Point-A-based planning using low dose rate and pulsed dose rate brachytherapy. Furthermore, local relapse-free survival was improved by an absolute 4% to 8% with the use of 3D planning in each of the three evaluated patient groups.46 To aid in the global adoption of 3D brachytherapy planning, the international IBS-GEC ESTRO-ABS group has established consensus CT-based guidelines for brachytherapy planning.47 48 Improved target delineation using MRI-based planning allows safe dose escalation to the high-risk clinical target volume (HR-CTV). For practices with access to planning MRI simulations, the GEC-ESTRO working group has also established guidelines for MRI-based brachytherapy.49 50

Other practice-guiding conclusions regarding brachytherapy were drawn from a retrospective analysis (retroEMBRACE) and a prospective observational study (EMBRACE I) in which about two-thirds of patients were classified with early-stage disease. Overall treatment time less than 50 days was associated with improved local control. A pelvic dose of 45 Gy prior to brachytherapy was associated with fewer pelvic insufficiency fractures versus 50 Gy as well as less radiotherapy-related diarrhea. Reductions in brachytherapy dose to the bladder, rectum, bowel, and vagina have all yielded lower rates of acute and late morbidity. An IMRT plan as part of definitive chemoradiation for a FIGO IIA tumor is presented in Figure 2A,B, using a dose of 45 Gy in 25 fractions. The EMBRACE studies also elucidated the relationship between dose and tumor control, demonstrating that values of HR-CTV D₉₀ of 85 to 95 Gy results in 95% local control for targets ≤30 cc and 90% for targets >30 cc. The 5-year local control was 92% across all FIGO stages. Nodal control remained high at 93% and 5-year overall survival of 74% for all patients.46–48 The guiding principles of retroEMBRACE and EMBRACE I led to the development of the currently-accruing EMBRACE II protocol. EMBRACE II seeks to implement risk-adapted nodal management, standardized IMRT protocols, and strict MRI-based brachytherapy goals which led to low rates of genitourinary, gastrointestinal, and vaginal sequelae.49–51 Figure 3A,B represents an example of EMBRACE-style high dose rate brachytherapy planning using a tandem and ovoid implant after external beam radiation therapy with concurrent chemotherapy.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

External beam radiation therapy planning using intensity modulated radiation therapy (IMRT). (A) Axial IMRT plan. Color wash axial representation of an IMRT plan delivering 45 Gy in 25 fractions to the pelvis with concurrent cisplatin. Patient simulated with an alpha cradle. Full and empty bladder scans were used for simulation to ensure coverage of the entire cervix and uterine fundus which is seen anteriorly. Legend: red=gross tumor volume (GTV); blue=nternal target volume (ITV); green=planning target volume (PTV) using EMBRACE II guidelines. (B) Sagittal IMRT plan. Sagittal view of IMRT plan with intravaginal marker to identify vaginal canal. Legend: green=PTV.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

High dose rate (HDR) magnetic resonance imaging (MRI)-guided brachytherapy using tandem and ovoids after chemoradiation. (A) HDR brachytherapy planning, axial (first of two implants). Legend: axial view of the first of two HDR brachytherapy implants per the EMBRACE II protocol. The tandem is visualized entering the cervix and the ovoids are seen on either side (black). Two implants were performed, each delivering two fractions of 7 Gy to the D90 high-risk clinical target volume (HR-CTV) (28 Gy over four fractions). Isodose lines shown represent the first two fractions (14 Gy). Light green=700 cGy; yellow=1400 cGy; pink=2100 cGy; red=post-external beam radiation therapy (EBRT) residual gross tumor volume (GTV); orange=HR-CTV. (B) HDR brachytherapy planning, sagittal (first of two implants). Legend: sagittal view of HDR brachytherapy plan per the EMBRACE II protocol. The tandem (black) is visualized entering the cervix and extending toward the uterine fundus. Isodose lines shown represent the first two fractions (14 Gy). Red=post-EBRT residual GTV; orange=HR-CTV. Cumulative doses (EQD2): HR-CTV D90=90.5 Gy; GTV D98=95.3 Gy. Bladder D2cc=76.1 Gy; rectal D2cc=55.2 Gy; sigmoid bowel D2cc=50 Gy.