Article Text

Download PDFPDF

Management of oligometastatic ovarian cancer recurrence during PARP inhibitor maintenance
  1. Eleonora Palluzzi1,
  2. Claudia Marchetti1,2,
  3. Serena Cappuccio1,2,
  4. Giacomo Avesani3,
  5. Gabriella Macchia4,
  6. Maria Antonietta Gambacorta2,3,
  7. Fabrizio Cocciolillo5,
  8. Giovanni Scambia1,2 and
  9. Anna Fagotti1,2
  1. 1 Dipartimento per la Salute della Donna, del Bambino e di Sanità Pubblica, Fondazione Policlinico Universitario Agostino Gemelli-IRCCS, Rome, Italy
  2. 2 Università Cattolica del Sacro Cuore, Rome, Italy
  3. 3 Dipartimento di Diagnostica per Immagini, Radioterapia Oncologica ed Ematologia, Fondazione Policlinico Universitario Agostino Gemelli-IRCCS, Rome, Italy
  4. 4 Dipartimento Servizi e Laboratori, Direzione Scientifica, Gemelli Molise Hospital, Radiotherapy Unit, Università Cattolica del Sacro Cuore, Campobasso, Italy
  5. 5 Nuclear Medicine Unit, Fondazione Policlinico Universitario Agostino Gemelli-IRCCS, Rome, Italy
  1. Correspondence to Dr Anna Fagotti, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Roma, Italy; anna.fagotti{at}


Objective The benefit of surgery and maintenance treatment with PARP inhibitors (PARPi) has been clearly demonstrated in ovarian cancer. Also, the efficacy and safety of stereotactic body radiotherapy has been shown in patients with metastatic, persistent, and recurrent disease. The aim of this study is to evaluate the management of oligometastatic progression during PARPi maintenance treatment.

Methods This is an observational, retrospective, single-arm study conducted from June 2017 to December 2020 in patients with recurrent ovarian cancer with oligometastatic progression under PARPi maintenance treatment and receiving surgery or stereotactic body radiotherapy for such recurrence. PARPi treatment was continued until further progression of the disease. The primary objective of the study was the median prolongation of the treatment-free interval-p (without platinum) after local treatment.

Results A total of 186 patients with ovarian cancer were treated with PARPi at recurrence. Of these, 30 (16%) developed oligometastatic progression. The median age was 49.5 years (range 35–73). Olaparib, niraparib and rucaparib were administered to 33%, 60%, and 7% of patients, respectively. The median prolongation of the treatment-free interval-p of patients treated with surgery or stereotactic body radiotherapy was 6 and 10 months, respectively (p=0.53). The median treatment-free interval-p of patients treated with surgery or stereotactic body radiotherapy at the time of oligometastatic progression was 32 and 29 months, respectively (p=0.44). At the time of this publication, 50% of patients are still on treatment with PARPi following progression.

Conclusions Patients with recurrent ovarian cancer who have oligometastic progression during PARPi maintenance may continue to benefit from PARPi if combined with local treatment.

  • ovarian neoplasms
  • cytoreduction surgical procedures
  • radiotherapy
  • BRCA1 protein

Data availability statement

We will provide our data for the reproducibility of this study in other centers if such is requested. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • The current management of oligometastatic progression in recurrent ovarian cancer is unclear.

  • The median prolongation of the treatment-free interval employing local therapy such as radiotherapy or surgery is 9 months.

  • Continuing PARP-i maintenance therapy in patients with recurrent ovarian cancer could be an option when oligometastatic disease progression is controlled by local therapy.


In about 70% of patients with advanced ovarian cancer the disease recurs within 5 years following first-line treatment.1 2 Traditionally, the management of recurrent ovarian cancer has been based on the length of platinum-free interval to define the treatment plan. Currently, this paradigm has been revised due to the use of biological agents such us PARP inhibitors (PARPi) and angiogenesis inhibitors to treat ovarian cancer in the first-line setting and beyond.1 Therefore, the treatment-free interval has been introduced as a relevant point in some clinical studies to define options for rechallenge with platinum-based or non-platinum-based chemotherapy and a surrogate of quality of life.1 3 4

The role of secondary cytoreductive surgery in the management of recurrent ovarian cancer has been controversial for many years. Primarily, the randomized GOG 213 trial showed that secondary surgical cytoreduction followed by chemotherapy did not result in longer overall survival than chemotherapy alone.5 The latest data from the DESKTOP III trial have shown a significant benefit of secondary cytoreductive surgery, in terms of overall survival, for patients with platinum-sensitive recurrent ovarian cancer, with a positive Arbeitsgemeinschaft Gynäkologische Onkologie (AGO) score and achieving complete gross resection.6 The advantage of secondary cytoreductive surgery in terms of progression-free survival was also shown in the recent results of the SOC-1 trial.7

Locoregional radiotherapy is also an option for treating recurrent ovarian cancer, especially in case of oligometastatic disease (vaginal/rectal implants, localized nodal or extranodal recurrence in the abdomen or pelvis).8–17 Stereotactic body radiotherapy in the management of metastatic, persistent, and recurrent ovarian cancer has shown a favorable toxicity profile, with complete and partial response rates of 65.2% and 23.8%, and a progression-free survival rate of 15.4% and 12.7% at 24 and 36 months, respectively.18–21 Patients with platinum-sensitive recurrent ovarian cancer are eligible to receive maintenance treatment with PARPi, if achieving a complete or partial response after platinum-based chemotherapy.22–26 Very few data are currently available about the possibility of repeating PARP inhibition in patients who were previously treated with PARPi as first-line treatment or following progression. The results of the AGO OVAR-2.31 (OReO) study, recently presented at the 2021 European Society for Medical Oncology (ESMO) Congress, showed that rechallenge with maintenance olaparib after platinum-based chemotherapy response resulted in longer progression-free survival for patients with relapsed ovarian cancer, regardless of their BRCA status. The median progression-free survival improved to 4.3 months in patients with BRCA mutant ovarian cancer and 5.3 months in the non-BRCA mutant cohort.27

On the basis of this background, we aimed to determine the survival outcomes in patients who developed oligometastatic progression while on PARPi and underwent either surgery or stereotactic body radiotherapy for local treatment followed by rechallenge with PARPi without platinum.


Study Design and Objectives

This is an observational, retrospective, monocentric, single-arm study including patients treated at the Ovarian Cancer Unit of Fondazione Policlinico Agostino Gemelli-IRCCS. The institutional review board approved the study (DIPUSVSP-PD-03–229) and patients gave informed consent for the collection and use of their data for a scientific purpose. In line with clinical practice, all recurrent ovarian cancer patients with progression during PARPi maintenance were discussed at the local multi-disciplinary tumor board. Progression was assessed every 4 months by either computed tomography (CT) scan or positron emission tomography (PET)/CT scan, according to RECIST criteria (version 1.1) or PERCIST criteria. Oligometastatic progression was defined as isolated disease progression (one lesion), discrete diffusion (up to five lesions in different locations),28 or progression of disease in ‘sanctuary’ sites such as brain or bone.

Patients whose oligometastatic disease was treated with either cytoreductive surgery or stereotactic body radiotherapy were included in the study. Cytoreductive surgery was offered in cases with a positive AGO score and abdominal disease only. All variables related to surgery were registered. Selected patients underwent stereotactic body radiotherapy, either because of lesion location or because they refused or were unfit for surgery. PARPi discontinuation occurred 5 days before and after any of the treatments. PARPi maintenance was resumed after surgery or radiotherapy and then continued until further progression of the disease and PARPi was not changed after local therapy. No patient enrolled in any protocol was included. The primary objective of the study was the prolongation of the treatment-free interval-p (without platinum) after local therapy. Also the treatment-free interval-p (from the last platinum treatment), defined as the time elapsed from the start of PARPi and the new chemotherapy line, was registered. As all patients waited an average of 4 weeks (range 4–5) before starting PARPi after platinum chemotherapy, this time was not included in the treatment-free interval-p. We will provide our data for the reproducibility of this study in other centers if such is required.


Patients were treated by linear accelerator or by MRIdian (ViewRay, Cleveland, Ohio, USA), a radiotherapy hybrid system that joins radiation delivery units (three cobalt sources) with magnetic resonance imaging (MRI) scanners. The set-up of patients was according to lesion site and treatment gating protocols used. The CT slice thickness was <3 mm to guarantee accuracy in delineation of the target. The use of intravenous contrast and fiducial markers was allowed. For all lesions the clinical target volume was defined as the visible tumor on diagnostic imaging. A further individually defined margin for each lesion was added to consider the organ motion. Volumetric modulated arc or step-and-shoot intensity modulated radiotherapy techniques were carried out based on the treatment machine. According to the target sites, the stereotactic body radiotherapy dose ranged between 25 Gy and 50 Gy delivered to all active metastatic lesions as per diagnostic imaging using a 5-day treatment schedule. Image-guided radiotherapy was used throughout all treatment fractions. Follow-up and monitoring of side effects was carried out with imaging and clinical evaluation every 4 months, at the same time as PARPi assessment of response.

Statistical Analysis

Medical, surgical, and biological data were retrieved from the REDCap (Research Electronic Data Capture) platform which is prospectively filled out and updated for all patients with ovarian cancer treated at our institution. Radiotherapy information was obtained directly from clinical charts. Every association was tested exclusively from a ‘hypothesis-generating’ perspective. Medians and life tables were computed using the product limit estimate by the Kaplan–Meier method, and the log-rank test was used to assess statistical significance. Statistical analysis was performed using IBM Microsoft SPSS version 85.0 for Mac (IBM Corporation, Armonk, New York, USA).


As shown in the consort diagram in Figure 1, from June 2017 to December 2020 a total of 1160 patients with recurrent ovarian cancer were treated and followed at our center. Of these, 186 (16%) women received PARPi maintenance treatment in a recurrence setting and outside any clinical trial, and 30 (16%) developed oligometastatic progression. Ten of these patients (33%) underwent cytoreductive surgery while 20 (67%) received stereotactic body radiotherapy. After locoregional treatment, all patients continued PARPi maintenance therapy and, at the time of data analysis, PARPi treatment is still ongoing in 15 patients (50%).

Figure 1

Consort diagram of the study. FUP, follow up; OMP, oligometastatic progression; ROC, recurrent ovarian cancer; SBRT, stereotactic body radiotherapy.

Patient features are shown in Table 1. The median age was 49.5 years (range 35–73). Twenty-nine patients had high-grade serous ovarian cancer and one patient had carcinosarcoma. Regarding mutational status, 16 (53%) patients were BRCA wild type (BRCA-wt), 10 (33%) had BRCA1 mutation, two (7%) had BRCA2 mutation, and two (7%) had a variant of uncertain significance (VUS) in the BRCA2 gene. Homologous recombination deficiency (HRD) status was not available for patients who were BRCA wild type or VUS. Eighteen (60%) patients were treated with niraparib, 10 (33%) with olaparib, and two (7%) with rucaparib. Before starting PARPi maintenance, 21 (70%) patients were at their first recurrence, seven (23%) patients were at their second recurrence, and two (7%) patients were at the third. No patients in first-line treatment were included in the study.

Table 1

Patient features

Fourteen patients (47%) underwent secondary cytoreductive surgery at the time of recurrence before starting platinum-based chemotherapy and then PARPi maintenance treatment. There was no residual tumor at secondary surgery in 12 of the 14 patients (86%) and residual tumor ≥1 cm in two patients (14%). Twenty-three (77%) patients received two lines of chemotherapy and seven (23%) were treated with more than two lines of chemotherapy. The median duration of PARPi administration until oligometastatic progression was 16 months (range 3–35) in the entire cohort.

Twenty-seven (90%) patients had a single site of oligometastatic progression while three (10%) had multiple sites of progression. The sites of oligometastatic progression were as follows: 17 (57%) lymph nodal (both supra- and sub-diaphragmatic), 10 (33%) parenchymal (including brain and bone), and seven (23%) peritoneal. Based on multi-disciplinary tumor board decisions, 10 (33%) patients underwent cytoreductive surgery while 20 (67%) patients were treated with stereotactic body radiotherapy. None of the patients had residual tumor after surgery. After stereotactic body radiotherapy, a complete response was registered in five (25%), a partial response in five (25%), and stable disease in eight (40%) patients. Two (10%) patients were not evaluable in terms of response at the time of data analysis. Only one G3 neurological complication was registered after neurosurgery, related to the excision of a right rolandic lesion requiring walking with crutches due to left hemiplegia, and two G3 side effects were observed after radiotherapy consisting of severe dysphagia requiring parenteral nutrition for 2 weeks and pericardial effusion requiring hospitalization and intravenous antibiotics (Table 2).

Table 2

Complications according to treatment received (surgery or stereotactic body radiotherapy)

After a median follow-up of 25 months (range 6–72) after the start of PARPi maintenance treatment and a median time of PARPi treatment before oligometastatic progression of 11 months (range 3–35), 15 (50%) patients had extensive progression requiring discontinuation of PARPi and treatment change. Of these, 40% had undergone surgery and 60% had undergone stereotactic body radiotherapy (Figure 1). Twenty-eight (93%) patients were still alive at the time of submission of this manuscript. In the overall population, median time from oligometastatic progression to further progression or last follow-up was 9 months (95% CI 5.8 to 12.16) (Figure 2A). The median prolongation of treatment-free interval-p of patients treated with surgery or stereotactic body radiotherapy for oligometastatic progression during PARPi was 6 and 10 months, respectively (p=0.53) (Figure 2B). The median treatment-free interval-p of the whole population was 32 months (95% CI 23.8 to 40.1) and the median overall treatment-free interval-p of patients treated with locoregional surgery or stereotactic body radiotherapy was 32 and 29 months, respectively (p=0.44). Due to the small sample size, we did not perform a multivariable model so it was not possible to identify any prognostic factors.

Figure 2

(A) Time from oligometastatic progression to subsequent progression in the overall population. (B) Time from oligometastatic progression to subsequent progression comparing surgery versus radiotherapy.


Summary of Main Results

Our study shows that local treatment (surgery or stereotactic body radiotherapy) of oligometastatic progression in patients with recurrent ovarian cancer under PARPi maintenance treatment is able to prolong the median overall treatment-free interval-p by up to 32 months. The rationale for such an approach is based on the supposed PARPi resistance developed in some clones from the original tumor growing as single or discrete nodules, whereas the rest of the disease is still under control. Although the integration of different therapeutic strategies to manage oligometastatic disease is quite common in other tumors such as non-small cell lung cancer, colorectal, and prostate cancer,29 there are no published data in the setting of recurrent ovarian cancer to date.

Results in the Context of Published Literature

The role of tertiary cytoreductive surgery is debatable,30–35 but it has shown a survival benefit in selected patients achieving complete gross resection.35 In our study, 40% of patients were treated with surgery at second recurrence and all had no residual disease. While surgery was mainly offered to patients with peritoneal recurrences, stereotactic body radiotherapy was used to treat nodal recurrences. Indeed, in the literature, nodal disease is significantly associated with a higher probability of achieving a complete response with stereotactic body radiotherapy.21 Interestingly, both local treatments (surgery and radiotherapy) had few severe complications and no differences in terms of survival benefit. Based on our results, it seems that each treatment can be offered based on the center’s availability, patient preference, and location of the disease.

Our results also underline the need for adequate follow-up to identify oligometastatic progression, avoiding delays that might hamper the efficacy of the local approach, either surgery or radiotherapy. The distribution of recurrence was in favor of lymph nodal recurrence (57%) with respect to the expected peritoneal (23%) and parenchymal (33%) disease (three patients had more than one site of recurrence). With the small sample size, it is difficult to clarify whether this is an effect of PARPi treatment or BRCA mutational status, or simply a selection bias related to the retrospective nature of the study. However, the rate of BRCA mutated patients in the lymph nodal recurrence group was 41%.

Oligometastatic disease is often asymptomatic, so four potential scenarios are feasible: (1) watch and wait, and treat with chemotherapy in case of symptoms; (2) perform only local treatment followed by observation; (3) standard chemotherapy; or (4) local therapy plus PARPi. We cannot compare our results with other strategies including chemotherapy instead of local treatment or observation waiting for symptom occurrence. The population of patients with oligometastatic progression after PARPi is emerging and further data are needed before drawing any definitive conclusion. To date, only data on patients with BRCA mutation continuing PARPi after progression and without adding any additional local treatment are available, reaching a median duration of post-progression PARPi treatment of 3.2 months.36 However, with this strategy tailored on oligometastatic progression, the median treatment-free interval-p increased by 9 months (95% CI 5.8 to 12.16) in our patients, thus significantly delaying the restart of another cytotoxic treatment.

Our patients had a median time of treatment with PARPi of 11 months (range 3–35) before oligometastatic progression, which is in line with registration studies showing a median progression-free survival of PARPi maintenance therapy ranging from 5.2 to 21 months, according to BRCA mutational status and timing of treatment.22–26 Results from the phase III ARIEL4 trial have shown that rucaparib significantly improves progression-free survival (median 7.4 months) compared with chemotherapy (median 5.7 months) in women with recurrent ovarian cancer carrying a BRCA mutation.37 At the American Society of Clinical Oncology (ASCO) Meeting 2021, a retrospective evaluation of clinical outcomes of subsequent platinum chemotherapy versus non-platinum chemotherapy in 29 patients progressing under PARPi maintenance treatment showed a median progression-free survival of 7.0 versus 8.5 months.36 These values are in line with those reported in our series, while adding locoregional treatment and continuing PARP inhibition. Relative to the median treatment time with PARPi, in the study by Brann et al38 this value was 9.0 and 5.5 months for platinum and non-platinum-based chemotherapy arms. The longer median treatment time with PARPi until oligometastatic progression in our series could be due to the high percentage of patients undergoing secondary cytoreductive surgery (47%) with complete cytoreduction (86%) before starting PARPi maintenance therapy, which has been shown to improve the effectiveness of PARPi as both first-line39 and second-line treatment.22–26

Strengths and Weaknesses

One of the main limitations of this study is the small number of cases, but our sample size would have allowed estimatation of the percentage of progression-free survival at 3 months, expected to be approximately 50% based on previous published data,36 with a Standard Error (SE) of 8%. Interestingly, in our series we exceed these data, with more than 75% of patients having definitive recurrence after more than 3 months. Due to the small series, a major limitation of the study is the absence of a comparison group of patients who developed oligometastatic recurrence while on PARPi treatment and who underwent either surgery or radiotherapy and stopped PARPi treatment. Also, molecular analysis of the metastases treated by surgery or stereotactic body radiotherapy would have improved the power of our analysis, but unfortunately this was not available. Finally, there is a heterogeneity of patients regarding the BRCA mutational status and the type of PARPi employed.

Implications for Practice and Future Research

Continuing PARPi maintenance treatment in addition to local therapy could be an option for patients with recurrent ovarian cancer who develop oligometastatic progression. This strategy allows for a delay in restarting a new cytotoxic platinum-based chemotherapy. Future research should focus on mutational analysis of surgically resected lesions to understand the molecular pattern of oligometastatic disease progression, such as PARPi resistance, and to propose a tailored target therapy.


Our findings show that patients treated with maintenance PARPi who develop oligometastatic progression may be safely treated either with stereotactic radiation or surgery followed by rechallenge with PARPi. The median time to subsequent disease progression is 9 months. Translational research will help elucidate the molecular biology of the disease and better understand who will benefit from PARPi rechallenge.

Data availability statement

We will provide our data for the reproducibility of this study in other centers if such is requested. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by DIPUSUVSP-PD-03-229. Participants gave informed consent to participate in the study before taking part.



  • Twitter @annafagottimd

  • Contributors All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by EP, SC and CM. The first draft of the manuscript was written by EP. The section on Radiotherapy was written by GM. Statistical analysis was performed by CM. Revision and validation of the manuscript was performed by AF, CM and GS. AF is the guarantor for the study. All authors read and approved the final manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests CM reports fees for advisory work from GSK, AstraZeneca, Pharmamar, and Clovis outside the submitted work. GS reports consulting fees or speaker fees from TESARO Bio Italy Srl, Clovis Oncology Italy Srl, Johnson & Johnson, AstraZeneca/MSD, Covidien AG, a Medtronic company, and from Baxter Healthcare SA outside the submitted work. AF reports consulting fees or speaker fees from Pharmamar, Johnson & Johnson, AstraZeneca/MSD and MSD Italia Srl outside the submitted work.

  • Provenance and peer review Not commissioned; externally peer reviewed.