Defining the role of high-dose radiation in oligometastatic & oligorecurrent cervical cancer.

Around 5-8 per cent of women diagnosed with cervical cancer present with metastatic disease at presentation and 16-25 per cent of patients fail at either within irradiated fields or at distant sites post-curative therapy in advanced cervical cancers. Conventionally, chemotherapy with palliative intent constituted the mainstay of treatment with modest survival outcomes and radiation therapy was reserved for symptomatic benefit only. While targeted therapies and immunotherapy have been added in therapeutic armamentarium, the impact on the outcomes is modest. In limited metastatic disease, radiation therapy to metastatic sites from different primary cancers has shown survival benefits; however, the data are scarce in cervical cancer. With a better understanding of the molecular biology of the metastases and recurrence pattern, emphasis is laid upon total eradication of the disease rather than offering relief from symptoms. This article summarizes the role of radiation therapy in limited metastatic disease and recurrent cervical cancer.

Cancer of the cervix is one of the most frequently diagnosed cancers among women. Every year, 575,000 women are diagnosed with invasive cervical cancer globally, with 13,800 in the United States, 54,500 in Europe and 96,922 in India with 65-75 per cent presenting in locally advanced stage12. The incidence of metastatic disease at the time of diagnosis ranges from five to eight per cent234. The 10 yr actuarial incidence of distant metastases ranges from 26, 39 and 75 per cent in International Federation of Gynaecology and Obstetrics (FIGO) stage IIB, IIIB and IVA, respectively56. Further, the most frequent sites of distant metastases are lung (21-39.3%), para-aortic lymph nodes (PALN) (11%), bone (16.3%), liver (12.2%), abdominal cavity (8%), brain (1.4%) and supraclavicular node (SCLN) (7%)78. Patients treated with concurrent chemoradiation (CCRT) and brachytherapy (BT) constitute a vast majority of patients who develop disease relapse at distant sites. The disease-free survival (DFS) in FIGO stage IIB-IV is 61-76 per cent, suggesting that close to 25-35 per cent of patients will present with progression9. Post-treatment failure is observed at distant sites in 16-25 per cent of patients10111213.

Chemotherapy (CTh) administered with palliative intent constitutes the mainstay of treatment in metastatic or recurrent setting with dismal survival of 2-18 months1415161718192021. Recently, metastatic disease or recurrent disease has been classified based on number of lesions, sites of involvement; with limited number (usually <3) and involvement as oligometastatic disease (OMD)22. Niibe and Hayakawa23 have suggested that if OMD is eliminated, a patient may be cured, as occurs in loco-regional tumours. A phase II trial by Palma et al24 utilizing stereotactic body radiation therapy (SBRT) in addition to systemic CTh in OMD reported overall survival (OS) benefit in the SBRT arm with grade 2 or higher toxicity 29 per cent with 4.5 per cent mortality in the SBRT arm versus nine per cent and no mortality in the standard of care arm. In addition, the updated analysis has shown durable response and no detrimental effect on the quality of life (QOL) of the patients25. Although the study had multiple primary cancer subtypes with OMDs, only a limited number of patients had gynaecological cancer and no patients with cervical cancer were included.

An international coordinated effort (ESTRO-ASTRO consensus) is in the process to better define this patient population that may benefit from intensified treatment approaches26. OMD presently includes patients who present with ≤3 metastasis and >3-5 metastases which is largely independent of the primary tumour and metastases location26. ESTRO-EORTC group has further classified OMD considering the timing of presentation of metastases, receipt of any systemic therapy before appearance of lesion and response of metastases to the systemic therapy22.

This review summarizes the existing evidence for the use of RT in the treatment of OMD and ORD in cervical cancer.

Role of systemic therapy agents

Currently, the recommended standard first-line regimen for the treatment of metastatic cervical cancer is the combination of cisplatin and paclitaxel, which has shown mild to modest improvement18. In addition, other chemotherapeutic agents including topotecan1519 and gemcitabine17 have been investigated and have shown slight improvement. Further, addition of immunotherapy and targeted therapies such as bevacizumab, pembrolizumab and cediranib has shown promising results, with bevacizumab combined with standard CTh recommended as a first-line therapy in metastatic or recurrent cervical cancer162027.

Though the results of these therapies are encouraging, their cost, availability and storage in developing countries have been a challenge and a limiting step for access. A cost-effective analysis done by Klag et al28 suggested that the combination of cisplatin and paclitaxel was the most cost-effective regimen and the addition of bevacizumab, although providing survival benefit, was not sustainable. Phippen et al29 suggested that the addition of bevacizumab to CTh was not affordable.

Combining radiation with targeted agents and immunotherapy

Various trials are underway combining SBRT or hypofractionated high-dose RT therapy and immunotherapy (NCT03452332, NCT03277482, NCT03614949, NCT03312114, NCT03192059) in recurrent or metastatic cervical cancer as shown in Table I. The accrual of these trials is ongoing, and the results are awaited which will help to further define the management and the optimal dose and fractionation schedules of RT.

Table I

Ongoing clinical trials of radiotherapy and immunotherapy in cervical cancer

Clinical trial identifier	Diseases	Immunotherapy	Radiation therapy	Endpoint	Secondary endpoint
NCT03452332 Phase I	Recurrent or metastatic cervical, vaginal or vulvar cancers	Tremelimumab + durvalumab	SABR with 3 fractions separated by 48 h	AE	Response to treatment; PFS; OS; TTNT
NCT03277482 Phase I	Metastatic or unresectable endometrial, ovarian (ovarian epithelial, fallopian tube, primary peritoneal), cervical, vaginal or vulvar cancer	Tremelimumab and durvalumab	Hypofractionated short course (either 1 or 5 days)	MTD	ORR; LRR, LCR, ARR, RD; PFS, OS
NCT03614949 Phase II	Recurrent or metastatic cervical cancer	Atezolizumab q3w 1 week	SBRT with 24 Gy in 3 fractions	ORR	PFS; OS
NCT03312114 Phase II	Metastatic fallopian tube cancer, primary peritoneal carcinoma, recurrent epithelial cancer of ovary	Avelumab	Stereotactic treatment (e.g. SABR/SBRT)	ORR	OS; CR; TTP; median response duration
NCT03192059 Phase II	Advanced or refractory cervical cancer, endometrial carcinoma, or uterine sarcoma	Immunomodulators Vitamin D, aspirin, cyclophosphamide, and lansoprazole plus curcumin with pembrolizumab	EBRT 24 Gy in 3 fractions, a fraction every 28 h	ORR	Incidence of AE; best OR; PFS; OS

AE, adverse event; ARR, abscopal response rate; CR, complete response; LCR, local control rate; LRR, local-regional recurrence; MTD, maximum tolerated dose; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; RD, response duration; SABR, stereotactic ablative radiation therapy; SBRT, stereotactic body radiation therapy; TTNT, time to next treatment; TTP, time to progression; EBRT, external beam radiation therapy

Role of pelvic radiation in patients with de novo metastatic disease

In patients with limited metastatic disease at the time of first clinical presentation, integration of local therapy with systemic therapy has been used by various investigators which has shown to provide progression-free survival (PFS) and OS benefits. The primary goal of delivering local therapies is to eradicate the local disease which could translate into clinical and survival benefits. Stenger et al30 analyzed 3169 patients of upfront metastatic cervical cancer treated with CTh alone versus CTh and pelvic RT; the addition of pelvic RT demonstrated significant survival benefit (23.2 vs. 10.1 months). Further, the median survival was longer in patients receiving RT dose >45 Gy and those receiving brachytherapy along with external beam RT with benefit seen even in patients with distant and nodal metastasis. A retrospective analysis of 2838 patients confirmed survival benefit (19.2 months vs. 10.1 months) with local definitive RT with CTh as compared to systemic CTh and palliative RT31. Another retrospective analysis by Yin et al32 confirmed the benefit of definitive RT with CTh over palliative RT and CTh and observed that mortality was due to distant progression rather than local progression after definitive RT. European Society for Medical Oncology (ESMO), European Society of Gynaecological Oncology (ESGO) and European Society for Radiotherapy and Oncology (ESTRO) guidelines also recommend treating pelvis (gross disease with elective irradiation of immediate nodal level) with CTh in localized metastatic disease33. Local RT also helps to alleviate the symptoms and pain associated with the disease. In a survey conducted within the EMBRACE group, it was revealed that all the participants agreed on delivering local RT in OMD with 68.2 per cent preferring CCRT and BT while 31.8 per cent preferred adding systemic therapy34. There is evidence that local RT improves survival in metastatic cervical cancer at presentation as shown in above studies; though most of the data are retrospective and prospective studies are needed. Further, no robust evidence exists about the benefit of elective irradiation of nodal targets while delivering local RT in such settings.

Role of radiation therapy in patients with metastasis to distant nodal sites

The spread of cervical cancer is stepwise with involvement of pelvic lymph nodes first followed by PALN and then systemic organs35.

Supraclavicular nodal (SCLN) metastasis: The incidence of SCLN metastasis is approximately 1.5-8.6 per cent with or without PALN metastasis at presentation with a five-year OS rate of 16.5 per cent3637. In a study of 24 patients who had distant nodal metastases at presentation, patients receiving CCRT followed by BT had better PFS and OS and complete response (CR) rates as compared to those receiving CTh alone38. In a retrospective analysis, 25 patients with both para-aortic nodes (PALN) and SCLN metastases received RT to the PALN and left SCLN (59.4 Gy) and 50.4 Gy to the pelvis with platinum-based CTh concurrently followed by BT. The median OS of the patients was 32 months with 64 per cent experiencing grade 3 or more acute haematologic toxicity39. Another study done in seven patients reported five year OS of 57 per cent when RT was delivered to SCLN along with pelvic RT and CTh, although the rate of acute haematologic toxicity was 67 per cent with no chronic toxicities40. In another study of 38 patients treated with definitive RT to OMD sites of cervical cancer including SCLN, mediastinum, lung and PALN, the median OS was 50.7 months and PFS was 21.7 months with <3 per cent grade ≥3 toxicity41. In such patients, CCRT followed by CTh was feasible with acceptable late toxicity, although the acute haematologic toxicity was reported to be higher. However, given uncommon presentation and differences concerning treatment, no consensus exists about the RT dose to SCLN. In addition, whether the target volume should include the entire nodal chain or only the involved node is unclear, and more studies focussing on these aspects are needed.

Inguinal nodal metastasis: The incidence of inguinal nodal metastasis at diagnosis is <2 per cent42. There are no robust guidelines for the management of inguinal node metastases; most of the evidence is based on case reports and individual practices. Being close to pelvic RT fields, the majority of RT oncologists extend the RT fields to include the inguinal nodes with concurrent CTh, with RT doses up to 45 Gy to the pelvis with an additional 9-15 Gy boost are recommended for the involved inguinal nodes and the preferred CTh agent is cisplatin 40 mg/m2 weekly43. The role of neoadjuvant or adjuvant CTh or node dissection is not defined in such a setting.

Mediastinal nodes metastasis: The incidence of mediastinal nodes at diagnosis is rare and is under-reported. Positron emission tomography computed tomography (PET-CT) is useful in the diagnosis of mediastinal nodes along with thoracotomy or video-assisted thoracic surgery. SBRT is becoming an attractive modality of treatment in recent times and a retrospective analysis of 52 patients with 84 mediastinal and hilar nodes treated with SBRT median dose of 35 Gy (range 30-50) in five fractions improved OS to 31.7 months44. A total of nine per cent local failures were observed at two years and 11.5 per cent of patients experienced grade 3 or more toxicity; however, a vast majority of toxicities were transient with 1.9 per cent developing grade 5 toxicity (radiation pneumonitis). The authors concluded that SBRT to mediastinal and hilar lymph node metastases was feasible with acceptable toxicity.

Cervical cancer with visceral metastases

Cervical cancer with lung metastasis: Approximately 4.1-7.7 per cent of patients with cervical cancer develop lung metastasis745. The number of nodules, possibility of surgical resection (SR), time interval between the appearance of metastases and initial treatment and receipt of CTh affect the outcomes4647.

In selected patients who present with limited metastatic disease, SR or RT targeting metastatic sites with systemic therapy should be offered. In a retrospective review of 529 patients with 776 lesions, after lung metastasectomy, the ninety month survival rate was 30 per cent, however very few patients with good general condition and adequate pulmonary reserve could undergo surgery48. SBRT is a feasible approach to resection. There are limited studies on SBRT in pulmonary oligometastatic setting, and most of the evidence of SBRT in the lung was derived from stage I non-small cell lung cancer who were medically inoperable4950. Studies utilizing SBRT in lung metastases from various primary cancers in de novo metastatic or recurrent setting are shown in Table II. In dose-escalation studies, patients treated with 8 Gy ×5 fractions showed CR in 51 per cent and partial response in 33 per cent with only one patient experiencing grade 3 or more toxicity60. In another dose-escalation study, SBRT was delivered 20 Gy ×3 fractions in 38 patients with 63 lesions, the actuarial local control (LC) was 96 per cent at two years with a median OS of 19 months with only eight per cent experiencing grade 3 or more toxicity61. Hou et al62 treated 19 patients with cervical cancer with 29 lung metastases with 56-64 Gy in 7-8 fractions and the one-year LC and OS were 75.6 and 76.8 per cent with only one patient developing grade 3 pneumonitis. They concluded that SBRT was safe and efficacious and could be an alternative to surgery62. Patients with a good general condition, limited pulmonary metastases (three or fewer), adequate pulmonary function and potentially treatable extra-thoracic disease can be considered as suitable candidates for SBRT. Total dose, fractionation depends upon the location of the tumour and proximity to critical structures. Based on two studies6364, dose constraints for SBRT in lung primaries are shown in Table III.

Table II

Studies evaluating the role of stereotactic body radiation therapy in lung metastasis

Study	Number of patients	Number of lung lesions	Primary site	Dose	LC	Toxicity
Wulf et al51, 2004	41	51	All (majority lung)	26-30 Gy SF 30-36 Gy/3#	One year LC 80 per cent	Grade 2 pnemominits 3 per cent
Hof et al52, 2007	61	71	All (majority lung)	24-26 Gy SF	Two years PFS 73 per cent	G3 pnemonitis 5 per cent
Ricardi et al53, 2012	61	77	All (majority lung)	26 Gy SF 45 Gy/3#	Two years LC 89 per cent	G3 pneumonitis 1.6 per cent
Osti et al54, 2013	66	103	All (majority lung, rectal, breast)	22 Gy (central) 30 Gy (peripheral)	Two years LC 82 per cent	G3 pneumonitis 11.9 per cent
Filippi et al55, 2014	67	90	All	26 Gy	Two years LC 88 per cent	G3 pneumonitis 1.6 per cent >Chest wall toxicity-8.9 per cent
Wersäll et al56, 2005	58	117	RCC	30-40 Gy/3#/1 week	LC 90 per cent at median FU 37 months
Milano et al57, 2012	121	103	All (majority breast and CRC)	50 Gy/10# over 2 weeks	Two years LC 77 per cent	G3 lung in 1 patient
Kang et al58, 2010	59	18	All (majority CRC)	39-51 Gy/3#	Three years LC 66 per cent	G1-2 pneumonitis in 46 per cent
Salama et al59, 2011	61	41	All	24-48 Gy/3#	Two years LC in 66 per cent	G3 pneumonitis in 1 patient

LC, local control; CRC, colorectal cancer; RCC, renal cell carcinoma; Gy, gray; SF, single fraction, #: fractions

Table III

The recommended dose volume constraints for stereotactic body radiation therapy in early stage non-small cell lung cancer

Parameters	RTOG 0236 protocol63	RTOG 0915 protocol64
Dose prescription	60 Gy/3#	34 Gy/1#	48 Gy/4#
		PTV	95 per cent PD to 95 per cent volume
		99 per cent PD to 90 per cent volume	95 per cent PD to 95 per cent volume
		99 per cent PD to 90 per cent volume	95 per cent PD to 95 per cent volume
			99 per cent PD to 90 per cent volume
CTV	100 per cent PD to 100 per cent volume	100 per cent PD to 100 per cent volume	100 per cent PD to 100 per cent volume
Spinal cord	Max <18 Gy	Max <14 Gy <0.35 cm3-10 Gy <1.2 cm3-7 Gy	Max 26 Gy <0.35 cm3-20.8 Gy <1.2 cm3-13.6 Gy
Lungs	V20 <10-15 per cent	<1500 cm3-7 Gy <1000 cm3-7.4 Gy	<1500 cm3-11.6 Gy <1000 cm3-12.4 Gy3
Heart	Max <30 Gy	Max <22 Gy 15 cm3 <16 Gy	Max <34 Gy <15 cm3-28 Gy
Oesophagus	Max <27 Gy	Max <15.4 Gy 5 cm3<11.9 Gy	Max <30 Gy
Proximal bronchial tree	Max <30 Gy	Max <20.2 Gy 4 cm3 <10.5 Gy	Max <34.8 Gy <4 cm3-15.6 Gy
Skin	Max <24 Gy 10 cm3 <40 Gy	Max <26 Gy 10 cm3 <23 Gy	Max <36 Gy <10 cm3-33.2 Gy
Brachial plexus	Max <24 Gy	Max <17.5 Gy 3 cm3 <14 Gy	Max <27.2 Gy <3 cm3-23.6 Gy

Superscript numerals denote reference numbers. PD, prescription dose; Gy, gray; PTV, planning target volume; CTV, clinical target volume; RTOG, Radiation Therapy Oncology Group

Cervical cancer with liver metastases: The incidence of cervical cancer with liver metastases is 1.2-2.2 per cent, with poor survival with CTh alone65. In limited liver metastases, resection of metastases is traditionally the choice of treatment with the majority of evidence comes from colorectal cancer and has shown excellent outcomes666768697071727374 as shown in Table IV. LC of liver metastases by use of SBRT is promising, providing 60-90 per cent at two years; however, tumour volume, receipt of prior CTh and RT dose have a definite role to play68697071. In spite of good local LC in treated site, distant progression is the cause of mortality; hence, combining systemic therapy with SBRT is justified; however, the sequencing of these therapies is crucial for adequate tumour control and survival. Severe toxicity related to SBRT is uncommon with the risk of RT-induced liver disease reported in SBRT is low75.

Table IV

Studies evaluating stereotactic body radiation therapy in liver metastases

Study	Lesions	Patients	Primary	Dose	LC	Survival	Toxicity
Blomgren et al69, 1995	Variable	31	Mixed	8-66 Gy/1-4#	80 per cent	NR	Haemorrhagic gastritis in 2 patients
Hoyer et al70, 2006	1-6 cm (<6)	44	Majority CRC	45 Gy/3#	Two years 86 per cent	Two years 62 per cent	Liver failure 1 Gastritis 2
Rusthoven et al67, 2009	1-3 (<6 cm)	47	Majority CRC	60 Gy/3#	Two years 92 per cent	Median 17 months	Grade 3 <2 per cent
Lee et al71, 2009	Variable	68	Majority CRC	28-60 Gy/3#	One year 71 per cent	Median 18 months	Grade 3-8 patients Grade 4-1 patients
Goodman et al72, 2010	1-5 (<5 cm)	26	Majority CRC	18-30 Gy/1#	One year 77 per cent	OS One year 62 per cent Two years 49 per cent	Grade 2-4 patients
Rule et al73, 2011	1-5	27	Majority CRC	30 Gy/3# 50-60 Gy/5#	One year 30 Gy 59 per cent 50 Gy 89 per cent 60 Gy 100 per cent	Two years OS 30 Gy 56 per cent 50 Gy 67 per cent 60 Gy 50 per cent	No grade 3 or more tox
Mahadeva et al74, 2018	Variable	427	Majority CRC	45 Gy/3# (range 12-60 Gy)	One year 84 per cent Two years 72 per cent	One year 74 per cent Two years 49 per cent	NR

Superscript numerals denote reference numbers. CRC, colorectal cancer; Gy, gray; OS, overall survival; LC, local control; NR, not reported; #: fractions

In a study evaluating the role of SBRT in various OMD sites (lung, liver and nodes) in 45 patients (9 patients with cervix primaries) with 70 lesions, the CR was 64 per cent with no patients progressing after achieving CR at a median follow up of survivors of 40 months with 13 per cent grade 1-2 acute toxicity and no grade 3 or more acute or long-term toxicity; no progression was seen in patients who achieved CR76. A similar study of treating oligometastatic sites with definitive RT has shown improved survival 42 (95% confidence interval: 21-63) months with no local or in-field recurrence77. Studies using high dose per fractions schedule with the aim to deliver ablative doses have shown improved survival in OMD at presentation787980 and including recurrent disease post-curative therapy as well8182. Administration of CTh before SBRT has shown poor tumour control likely due to the killing of sensitive clones and remains of CTh-resistant clones83. This study predicted that in patients receiving no CTh before SBRT, biological effective dose (BED) of 209±67 Gy, but in those receiving CTh prior to SBRT, BED of 286±78 Gy needed for 90 per cent control probability at two years. In addition, BED >100 Gy, tumour volume <40 cm3 and metastasis with the head neck (median 37 months) and breast (32 months) primary tumours have better survival than arising from colorectal (30 months) and lung primaries (26 months)84. In a study by Hong et al85 in 89 patients treated with SBRT to liver metastases, the mutation in KRAS oncogene was the strongest predictor of poor LC and tumours with both KRAS and p53 mutation were radioresistant with one-year LC 20 per cent versus 69.2 per cent in the non-mutated cohort. This study highlights the importance of tumour genotyping before SBRT and treatment intensification in such a subset of patients85. Ideal candidates for liver metastasis SBRT should have a good performance status, sufficient hepatic reserve, no metastatic disease outside liver and an uninvolved liver volume of 700 ml or greater. Table V shows dose constraints from different studies for liver SBRT.

Table V

Recommended dose volume constrains for stereotactic body radiation therapy in liver metastases

Structures	Wulf et al75	Rusthoven et al67	Hoyer et al70	QUANTEC86
Prescription dose	Low dose group-3×10 Gy or 4×7 Gy prescribed to the PTV-encl 65 per cent isodose High dose group-3×12-12.5 Gy or 1×26 Gy/PTV enclosing 80 per cent isodose	12-20 Gy×3 fractions prescribed to isodose line covering PTV	15 Gy×3 fractions	NA
Liver-CTV	30 per cent <21 Gy 50 per cent <15 Gy	700 ml <15 Gy	700 ml <15 Gy	700 ml <15 Gy Dmean <15 Gy
Stomach	D5 ml <21 Gy	Dmax ≤30 Gy	D1 ml <21 Gy	Dmax <30 Gy
Bowel	D5 ml <21 Gy	Dmax ≤30 Gy	D1 ml <21 Gy	Dmax<30 Gy
Oesophagus	D5 ml <21 Gy	NA	D1 ml <21 Gy	NA
Bilateral kidney	NA	Dmax <18 Gy D35 <15 Gy	Dmax <18 Gy D35 <15 Gy	NA
Spinal cord	NA	Dmax ≤18 Gy	Dmax ≤18 Gy	Dmax ≤20 Gy
Heart	D5 ml <21 Gy	NA	D1 ml <30 Gy	NA

Gy, gray; PTV, planning target volume; NA, not applicable; CTV, clinical target volume; QUANTEC, quantitative analyses of normal tissue effects in the clinic; Dmax, maximum density

Cervical cancer with bone metastasis: Bone metastases incidence varies from 0.8 to 23 per cent in cervical cancer8788. Vertebral column, mainly the lumbar and thoracic spine (48%) followed by pelvis, is the most common site of involvement, with the majority of them (67%) detected within the first year of the radical treatment. Vertebral metastases, if left untreated or delayed, can cause spinal cord compression and irreversible neurological deficit. RT can provide pain relief and can stabilize fractures; however, the majority of patients can relapse87. In a study of 105 patients treated with an RT dose of 30 Gy in 10 fractions, the median survival was 10 months with 60 per cent of patients responding to pain. In addition, the use of local treatment was associated with improved survival than the survival of seven months observed in patients receiving CTh alone (P=0.011)89. In spine metastases, the traditional dose fractionation regimen used was 8 Gy single fraction, 20 Gy in five fractions and 30 Gy in 10 fractions providing good symptomatic relief but poor LC. With high-dose SBRT ranging from 15 to 45 Gy in 1-5 fractions, LC improves along with PFS and delay in CTh switchover90. In other histologies, bone SBRT has increased LC up to 85-90 per cent using 15-30 Gy in 1-3 fractions9192. In a retrospective analysis of 1400 patients, LC was 90 per cent at 15 months post-SBRT with <1 per cent risk of myelopathy93. In patients post-laminectomy, spine SBRT is feasible with one-year LC ranging from 85 to 95 per cent with no grade three or higher acute or late toxicities9495. In non-spine metastases, SBRT 30-35 Gy in five fractions showed excellent LC of 87 per cent at two years with 8.5 per cent fracture rates in the treated sites and the man time to fracture was 8.4 months96. Similarly, post 24 Gy single fraction, the LC was 91.4 per cent at one year with no late grade three or higher toxicities and pain resolution in 88 per cent of the patients with non-spine bone oligometastasis. Two patients developed pathological fractures, but both were asymptomatic97. Various studies9899, meta-analysis100 and ASTRO statement101 have confirmed that single-fraction RT therapy is as efficacious and safe as fractionated RT; however, the retreatments rates are higher with single-fraction RT. The addition of bisphosphonates and denosumab with RT has shown benefit in reducing skeletal-related events and combining them is prudent. Sprave et al102 assessed QOL in patients with spine metastases post-conventional RT and SBRT, and there was no difference between the two regimens across all domains and pain-related scores.

Cervical cancer with brain metastases: Brain metastasis in cervical cancer is rare ranging from 0.5 to 1.2 per cent103. Brain metastases cause not only morbidity and mortality but also neurocognitive decline, leading to poorer QOL. Good prognostic factors are age less than 50 at diagnosis, good performance status, single or less than three lesions and absence of extra-cranial lesions104.

With palliative whole-brain RT (WBRT), the survival ranges from 3 to 7 months104105. In single or limited metastasis, SR followed by WBRT has shown significantly better outcomes as compared to WBRT alone, 10-11.5 months versus six months106107108. Further, the LC rates and OS for patients with a single metastasis treated either with SR followed by WBRT or with stereotactic radiosurgery (SRS) alone are similar106107109110. WBRT followed by SR was beneficial to decrease local and distant recurrence, although there was no benefit in OS111.

SRS plus WBRT has shown benefit in OS (10.6 vs. 5.3 months, P=0.001), LC (88.9 vs. 55.6%) and time to progression (8.1 vs. 4 months) as compared to WBRT alone in brain metastasis in lung cancer113. Two studies have confirmed that SRS boost has local, survival benefit over WBRT alone in limited brain metastases and should be considered in patients with good performance status and controlled extra-cranial disease112113.

Studies evaluating the role of SRS in cervical cancer with brain metastases are limited; however, all three studies have shown survival and LC benefit114115116.

There is an evidence to suggest that SR when done before WBRT in a single metastasis versus upfront WBRT alone leads to improved OS and functional status in the patients106107. At present, there is no level I evidence justifying WBRT or SRS after surgery and the decision of adjuvant treatment should be made judiciously.

While LC and OS continue to be the prime end points, neurocognitive deterioration should be considered while planning treatment and emphasis must be given to improve the QOL of the patients. The role of SRS is evolving and currently limited to boost after WBRT, as monotherapy in <5 metastases, as salvage therapy after the previous WBRT, post-operative RT and in radio-resistant brain metastases.

Cervical cancer with peritoneal deposits: Peritoneal deposits in squamous cell carcinoma of the cervix at diagnosis are extremely rare and comparatively more in adenocarcinoma than in squamous histology. Conventionally, CTh with palliative intent used to be the treatment modality. One case report has shown that excision of limited deposits followed by CCRT (including the deposit and track site of excision) can be curative and provides long-term disease control and survival. However, no definitive guidelines exist for the treatment of peritoneal deposits and decisions should be made on an individual basis depending upon the patient`s conditions and extent of disease.

de Vin et al117 proposed an algorithm based on four risk factors affecting OS in OMD: presence of non-adenocarcinoma histology, presence of intracranial metastases, synchronous OMD and male gender. Based on this, the OS ranges from 40 to 4 months in the presence of 0 and all four risk factors, respectively117.

Locoregional recurrence after curative therapy

The incidence of nodal recurrence post-curative therapy in locally advanced cervical cancer ranges from 4.7 to 18.9 per cent while the central recurrence (local and regional) rate ranges from 7.6 to 25 per cent and isolated para-aortic ranges from 1.7 to 12 per cent101113118119. Patients presenting with only local relapse with no distant metastasis can be offered treatments with curative intent with the choice of treatment depending upon the receipt of the prior treatment. Women who are likely to benefit from surgical management include those who present with a central pelvic recurrence without sidewall fixation or associated hydronephrosis, and small tumour size. Patients who present with loco-regional relapse post-hysterectomy can be offered RT with or without CTh. Vaginal vault or parametrial relapse can be considered for external beam RT with BT or BT alone with five-year survival ranging from six to 55 per cent120121. Mahantshetty et al122 analyzed 30 patients treated with brachytherapy to a median dose of 42 Gy (range 37 to 46 Gy) and reported two-year LC 44 per cent, PFS 42 per cent and OS 52 per cent with grade three proctitis and cystitis and grade two small bowel toxicity in three (10%) patients. In addition, long intervals between two RT schedules and high brachytherapy dose favoured better outcomes122. In a phase III randomized study comparing neoadjuvant CTh followed by surgery versus RT and CTh in FIGO stage IB2, IIA and IIB, 12.3 per cent of patients developed local recurrence only and 6.3 per cent received local and distant post-surgery; around 30 per cent of them received salvage RT123. Although the primary analysis showed difference in DFS between the two arms, no difference was seen in OS. Effective salvage of recurrence using RT with or without CTh may be one of the reasons123. NCCN 2020124 and ESMO-ESGO-ESTRO33 guidelines state that patients with central recurrence, who have not received RT or failed outside the treatment fields, should undergo surgical resection if feasible followed by adjuvant RT (including brachytherapy) if feasible, and systemic therapy. In patients who have received prior RT and have a central recurrence, pelvic exenteration is advised, while in non-central disease, pelvic re-irradiation can be offered or systemic CTh33124.

In patients presenting with nodal relapse, surgical debulking, RT with CTh, or best supportive care can be offered; however, the prognosis is variable; with three-year OS rate of patients who underwent RT and CT being 85.7 per cent; surgery 66.7 per cent; CTh only 48.8 per cent; RT only 41.3 per cent and best supportive care 0 per cent (P=0.014)125. Re-irradiation with or without CTh is feasible without much side toxicity. In a study, 22 cervical cancer patients with LN recurrence post-surgery were treated with salvage RT (median dose 60 Gy) with (n=18) or without (n=4) CTh126. Patients treated with CTh and RT achieved a longer five-year PFS 72.9 per cent and OS rate 60 per cent with less than 20 per cent recurrence occurring inside the RT field126. In a retrospective analysis of 28 patients with recurrent genitourinary malignancies after a median disease-free interval of 9.5 years, RT to a median dose of 50 Gy using hypofractionated schedule showed good symptomatic relief and no grade 2-4 toxicity127. The authors suggested that a median cumulative dose of 100 Gy could achieve successful palliation without much toxicity in a patient previously treated with RT127. A clinical trial exploring the role of IMRT in the re-radiation of the pelvis in recurrent cervical cancer is ongoing (NCT03170570). SBRT is an attractive option; it can deliver higher doses using highly conformal RT and in a shorter overall treatment time. SBRT has been tried in para-aortic nodal recurrences from gastric, prostate and gynaecological primaries128129130131132133 as shown in Table VI. In a study on 91 patients (13% patients had cervical cancer primary) Loi et al134 treated pelvic and para-aortic nodal relapse with SBRT 48 Gy in six fractions (biological equivalent dose of 86 Gy), and showed a median OS of 36 months and PFS of 79 per cent at four years with no late grade three or more toxicity. Park et al82 treated 100 patients of recurrent oligometastatic cervical cancer using SBRT and reported two-year PFS of 82 per cent and OS of 57 per cent. Choi et al132 treated 30 patients using SBRT 33-45 Gy in three fractions with four-year LC and PFS of 67.4 and 45 per cent, respectively, with grade three toxicity occurring in one patient only after 20 months of treatment. Although the evidence in pelvic re-irradiation using hypofractionated SBRT in cervical cancer is emerging, data from other pelvic malignancies have shown promising results135. A survey within the EMBRACE network showed that for out-of-RT field nodal recurrences, 63.7 per cent preferred treating with the intent of curing the disease with RT and CTh, while for in RT field recurrences, palliation was the aim of the treatment34. Thus, re-irradiation of the pelvis in recurrent cervical cancer is an area of significant uncertainty and further trials are warranted.

Table VI

Select studies showing outcomes and toxicities of stereotactic body radiation therapy in para-aortic nodal recurrence

Author (year)	Number of patients	Re-irradiation	Primary cancer	SBRT dose	Outcomes	Toxicity
Bonomo et al128, 2013	26 (32 nodes)	Not reported	Gynaecological and prostate	36 Gy/3# in majority	LC-100 per cent Distant progression in 8 per cent All patients alive till last follow up	No acute or severe toxicity
Corvò et al129, 2013	33	3 ReRT (median previous RT dose-30 Gy)	Pancreas and colon	35 Gy/5# weekly	LC-83 per cent at 2 years 16 patients at median follow-up of 28 months	No acute or late grade 3 or more toxicity
Jereczek-Fossa et al130, 2012	69, (94 nodes)	20 lesions	Gastro-intestinal, prostate	24 Gy in 3#	Three years LC-64.3 per cent OS-49.9 per cent	Late grade 3 or more in 3 patients
Kim et al131, 2010	7	No	Gastric	48 Gy/3#	Three years OS-43 per cent, local relapse in 1 patient	No late toxicity
Choi et al132, 2009	30	4 patients	Cervix, endometrial, gastric	33-45 Gy/3#	Four years LC-67.4 per cent PFS-45 per cent OS-50.1 per cent	Late grade 3 in one patient
Bignardi et al133, 2011	19	No	Miscellaneous	45 Gy/6#	Two years LC-77.8 per cent PFS-19.7 per cent OS-93.3 per cent	Late grade 3 in one patient

OS, overall survival; PFS, progression-free survival; LC, local control; RT, radiation therapy; SBRT, stereotactic body radiation therapy; Gy, gray

Future considerations

The ESTRO-EORTC expert groups have classified OMD based on different characteristics of the patients who underwent treatment with curative intent and sub classified into oligorecurrence, oligoprogression and oligopersistence, which is being prospectively evaluated22. These authors have started OLIGOCARE project, a prospective project which is currently accruing patients with OMD, where the researcher can follow up their work which will establish further which treatment is best suited to individual patients (available from: ). A nomogram is also available to stratify oligometastatic patients based on the sex of the patient, timing of presentation of the disease, presence of intracranial metastases, histology and KPS score and to plan individualized care136. A recent survey highlighted the need of joint approach and clinical trials to decide the optimal management of OMD and ORD in cervical cancer34.

Conclusion

Metastatic cervical cancer possesses a challenge in diagnosis and treatment, and with conventional CTh, the survival remains poor. Treatment intensification using RT in local and metastatic sites, especially SBRT, has shown promising results with improved OS and PFS. As the role of SBRT continues to grow, the utility of this approach in cervical cancer needs to be explored.