Effectiveness of oral etoposide in recurrent or refractory epithelial ovarian cancer, primary peritoneal cancer and fallopian tube cancer

Abstract

The aim of treatment in recurrent or refractory epithelial ovarian cancer is palliation. In a patient with platinum-resistant status, several chemotherapy regimens have been reported with similar response rates. Among these agents, the oral etoposide holds an advantage of the route of administration and management in the out-patient setting. This retrospective study was conducted to evaluate the effectiveness of oral etoposide. Data of patients with recurrent or refractory epithelial ovarian, primary peritoneal and fallopian tube cancer who received oral etoposide treatment in Ramathibodi Hospital, Mahidol University from January 1997 to December 2017 were collected. Progression-free survival (PFS) and overall survival (OS) were primary and secondary outcomes, respectively. The oral etoposide at a dose of 50 mg/m² was prescribed. Sixty-six records were analysed. Median OS and median PFS were 8.3 months (95% confidence interval (CI): 6.8, 10.4) and 3.1 months (95%CI: 2.3, 3.9), respectively. Other factors including age, body mass index (BMI), histopathology subtype, primary treatment, result of the primary surgery, platinum status, site and size of recurrent cancer, treatment after discontinuation of oral etoposide and the line of chemotherapy regimen were not associated with the prognosis. The initial cancer stage was the only independent poor prognostic factor. The main toxicity was neutropenia.

Impact Statement

• What is already known on this subject? After the recurrence of epithelial ovarian cancer, peritoneal cancer or fallopian tube cancer, the aim of treatment in this setting is palliation with accentuating on symptomatic control and enhancing the quality of life. According to previous clinical trials, the chemotherapy regimens which were considered as a second-line or beyond the second-line therapy have been reported with similar response rates. Among these agents, the oral etoposide could be administrated by oral route. There were several articles reported the effectiveness of oral etoposide in different dosage.

• What do the results of this study add? This study adds that administration of oral etoposide at a dose of 50 mg/m² showed fairish oncologic outcomes with manageable toxicity.

• What are the implications of these findings for clinical practice and/or further research? The results will provide evidence that the oral etoposide can be considerate as a choice of palliative chemotherapy because of an advantage of the route of administration and management in the out-patient setting.

Keywords:

• Chemotherapy
• epithelial ovarian cancer
• oral etoposide
• platinum-resistant ovarian cancer
• recurrent ovarian cancer

Introduction

Recently reported, ovarian cancer is the eighth common female cancer and the eighth common cause of death among women worldwide (Bray et al. 2018). Ovarian cancers are classified based on the cell type from which they originate: surface epithelial, germ cells or sex cord-stromal cells. Approximately, 80–90% of ovarian cancers are of epithelial origin. According to the International Federation of Gynecologists and Obstetrics (FIGO) recommendation, primary surgical staging is the main treatment. Followed by adjuvant chemotherapy which accepted standard is six cycles of platinum-based combination chemotherapy, a platinum (carboplatin or cisplatin) and a taxane (paclitaxel or docetaxel) can be considered for the first-line chemotherapy after primary treatment or neoadjuvant treatment (NACT). Despite the optimal cytoreductive surgery and appropriate chemotherapy, more than 50% of the patients relapsed within 2 years (Piccart et al. 2003).

After recurrence, the aim of treatment in this setting is palliation with accentuating on symptomatic control and enhancing the quality of life. One of the clinical factors that must be taken in account when considering a second-line chemotherapy regimens is platinum status. A group of patients with recurrence after 6 months is defined as platinum sensitive, whereas platinum resistance is defined when the relapse happens during or within 6 months after the platinum therapy (Markman et al. 1991; Friedlander et al. 2011). Recurrent patients with platinum sensitive status who resist to the first-line chemotherapy also consider for the second-line treatment. According to previous clinical trials, the second-line or beyond second-line regimens, for instance, 20% of topotecan, 19% of gemcitabine, 26% of liposomal doxorubicin and 27% of oral etoposide, have been reported with similar response rates (Rose et al. 1998; Gordon et al. 2004; Ferrandina et al. 2008; Markman 2011). Among these agents, the oral etoposide holds an advantage of the route of administration and management in the out-patient setting.

Etoposide targets at DNA topoisomerase II, which is an enzyme causing transient double-strand DNA break. With the appearance of etoposide, the DNA topoisomerase II complex was stabilised resulting in inhibition of DNA rejoining (Ross et al. 1984; Rose et al. 1998). Etoposide has been studied in numerous types of cancer including testicular cancer, lung cancer, lymphoma and neuroblastoma (Fellows of the American Society of Health-System Pharmacists. 2016). In the late 1980s, etoposide was first noted to be active in epithelial ovarian cancer, and a variety of trials have confirmed its efficacy (Hoskins and Swenerton 1994; Ozols 1999; Baur et al. 2005; Thavaramara et al. 2009; Bozkaya et al. 2017; Kumar et al. 2018). The present role of etoposide is in the recurrent or refractory epithelial ovarian cancer, primary peritoneal cancer and fallopian tube cancer (Morgan et al. 2016).

We conducted this retrospective study to evaluate the effectiveness of oral etoposide in recurrent or refractory epithelial ovarian cancer, primary peritoneal cancer and fallopian tube cancer. Adverse effects and toxicity profiles will also be described.

Materials and methods

This is a retrospective study based on medical records and electronic medical records of recurrent or refractory epithelial ovarian cancer, fallopian tube cancer or primary peritoneal cancer. All patients who had been treated with oral etoposide in Ramathibodi Hospital during 20-year period from January 1997 to December 2017 were included. The patients with an incomplete medical record or synchronous malignancy were excluded from the study. The research protocol was approved by the Ethical Committee for Research involving Human Subjects of the Faculty of Medicine Ramathibodi Hospital, Mahidol University, Thailand. The protocol number was MURA 2018/232.

In our institute, one cycle of etoposide, 25 mg/capsule was administered at 50 mg/m² orally per day for 21 days with a seven-day break. A complete blood count (CBC), serum creatinine, liver function test was performed before the prescription of every cycle. Pre-treatment laboratory required absolute neutrophil count >1500/mm³, platelet count >100,000/m³, creatinine <2 mg/dL. The pre-treatment evaluation before every cycle also included history taking, physical examination, performance status assessment with Eastern Cooperative Oncology Group (ECOG) score and evaluation of side effect and toxicity. Toxicity was graded according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.03 (2010; CTCAE 2010). The patients who did not meet the prerequisite pre-treatment criteria were subjected to delay in treatment or dose reduction. The response of treatment was evaluated every three cycles or whenever clinically indicated. Evaluation Criteria In Solid Tumour (RECIST) version 1.1 was used for imaging investigation (Eisenhauer et al. 2009).

To assess the effectiveness of the oral etoposide, the primary outcome is progression-free survival (PFS), defined as an interval from the date of oral etoposide initiation to the date of disease progression or last follow up. Overall survival (OS), calculated from the date of oral etoposide initiation to the date of patients’ death from any cause or last follow up, was evaluated as a secondary outcome.

Statistical analyses were performed using STATA version 15.1 (StataCorp, College Station, TX). For continuous data, means and standard deviations (SDs) or medians (ranges) were provided according to the distribution of data. For categorical data, frequency and percentage were reported. Univariable and multivariable Cox regressions were used to estimate hazard ratio (HR) and 95% confidence interval (CI) for each interested factor and the PFS and OS. Factors that yielded p value <.1 using univariable analysis were included in the multivariable analysis. A p value of <.05 was regarded as statistically significant. The variables included in this analysis were age, body mass index (BMI), FIGO stage (stage I, stage II, stage III, stage IV or unclassified), type of cancer (ovarian cancer, primary peritoneal cancer or fallopian tube cancer), primary treatment (primary debulking surgery or NACT), residual disease after surgery (optimal (≤1 cm) or suboptimal (>1 cm)), histopathological subtype (clear cell carcinoma, serous carcinoma or others), the site of tumour recurrence (intra-abdomen or intra- and extra-abdomen), the largest size of recurrent tumour (3 cm or larger than 3 cm), numbers of chemotherapy regimens before oral etoposide (≤3 or >3 cycles), and treatment after discontinuation of oral etoposide (chemotherapy or best supportive care).

Results

During the studied period, a total of 68 cases were identified. Two patients were excluded due to incomplete medical records and synchronous malignancy, leaving 66 patients for analysis. The patient characteristics are given in Table 1. Mean age at cancer diagnosis was 54 years. The majority of patients were diagnosed with ovarian cancer followed by primary peritoneal cancer, but no fallopian tube cancer was reported. Approximately, 70% was classified as an advanced stage (FIGO stage III or IV). Serous carcinoma and clear cell carcinoma were common histopathologic subtypes in our cohort, accounting for approximately 35% for each cell type. Primary debulking surgery was performed on 57 patients (86.36%). Thirty-six patients (54.55%) achieved optimal surgery whilst 30 patients (45.45%) had suboptimal surgery.

Table 1. Patients’ characteristics.

	N (%)
Total number of patients	66
Age (years); mean±SD	54 ± 9.35
BMI (kg/m^2)
<18.5	7 (10.61)
18.5–22.9	25 (37.88)
>23	34 (51.51)
Type of cancer
Ovarian cancer	62 (93.94)
Primary peritoneal cancer	4 (6.06)
Fallopian tube cancer	0
FIGO stage
I	8 (12.12)
II	10 (15.15)
III	38 (57.58)
IV	9 (13.64)
Unclassified	1 (1.51)
Histology
Serous carcinoma	23 (34.85)
Clear cell carcinoma	23 (34.85)
Endometrioid carcinoma	5 (7.58)
Malignant Brenner tumour	4 (6.06)
Others	11 (16.66)
Primary treatment
Surgery	57 (86.36)
NACT	9 (13.64)
Optimal surgery
Yes	36 (54.55)
No	30 (45.45)
Site of cancer recurrence
Intra-abdomen	64 (96.97)
Intra- and extra-abdomen	2 (3.03)
Largest cancer size at recurrence
<3 cm	31 (46.97)
>3 cm	35 (53.03)
Line of oral etoposide
Second-line treatment	3 (4.55)
Third-line treatment	31 (46.97)
Fourth-line treatment	22 (33.33)
Fifth-line treatment	5 (7.58)
Sixth-line treatment	3 (4.55)
Seventh-line treatment	1 (1.52)
Eight-line treatment	1 (1.52)
Total cycle (median (range))	3 (1–24)

BMI: body mass index; NACT: neoadjuvant chemotherapy.

Recurrence mostly occurred in the intra-abdominal cavity (96.97%) with a size above 3 cm in 35 patients (53.03%). The oral etoposide, as a third-line treatment, was prescribed to 31 patients (46.97%). Median of oral etoposide cycles received was 3 (ranging from 1 to 24). Approximately, 60% of the patients discontinued oral etoposide because of progressive disease while 15 patients (22.7%) stopped as a consequence of intolerable adverse effects. Only one patient finished the medicine with a complete response. There were two patients who received palliative radiotherapy concurrently with oral etoposide. After the discontinuation of oral etoposide, 23 patients (34.85%) were treated with other chemotherapy regimens. The remaining patients were treated with the best supportive care. The median PFS and OS were 3.11 (95%CI: 2.30, 3.67) months and 8.89 (95%CI: 6.95, 10.66) months, respectively.

The result from univariate analysis is demonstrated in Table 2. For PFS, the FIGO stage was the only bad prognostic factor. Patients with greater BMI values at diagnosis had more favourable OS. The HR was 0.26 (95%CI: 0.11, 0.61, p = .002) and 0.36 (95%CI: 0.15–0.85, p = .019) if the BMI values were 18.5–22.9 kg/m² and ≥23 kg/m², respectively. The more advanced the FIGO stage was classified, the worse the OS was shown. Primary treatment with NACT seemed to be a bad prognostic factor for the OS when compared with primary treatment with upfront surgery with HR of 2.79 (95%CI: 1.31, 5.90, p = .007). Other factors, such as age, histopathology subtypes, residual tumour after surgery, the site and size of the tumour at recurrence, line of oral etoposide or treatment after discontinuation of oral etoposide, were not associated with the survival prognosis.

Table 2. Univariate Cox regression analysis for oncologic outcomes.

Factors	Progression-free survival			Overall survival
	HR	95%CI	p Value	HR	95%CI	p Value
Age	0.99	0.96–1.02	.521	0.99	0.96–1.02	.524
BMI (kg/m^2)
<18.5	1			1
18.5–22.9	0.61	0.26–1.42	.254	0.36	0.15–0.85	.019
≥23	0.54	0.24–1.25	.152	0.26	0.11–0.61	.002
FIGO stage
I	1			1
II	1.81	0.66–4.99	.252	1.24	0.49–3.15	.657
III	2.34	0.97–5.61	.057	1.54	0.71–3.35	.272
IV	3.80	1.33–10.89	.013	3.69	1.38–9.88	.009
Unclassified	2.09	0.25–17.64	.500	1.14	0.14–9.22	.902
Histology
Serous carcinoma	1			1
Clear cell carcinoma	0.76	0.44–1.32	.329	0.99	0.57–1.75	.996
Others	0.92	0.46–1.84	.816	0.97	0.48–1.93	.924
Primary treatment
Surgery	1			1
NACT	1.40	0.69–2.86	.355	2.79	1.31–5.90	.007
Optimal surgery
Yes	1			1
No	1.57	0.96–2.58	.074	0.88	0.53–1.45	.618
Site of tumour recurrence
Intra-abdomen	1			1
Intra- and extra-abdomen	0.17	0.02–1.26	.082	0.51	0.12–2.13	.360
Largest recurrent tumour size
≤3 cm	1			1
>3 cm	1.61	0.97–2.70	.068	0.93	0.57–1.52	.763
Line of etoposide
≤3rd line	1			1
>3rd line	1.06	0.63–1.76	.837	0.99	0.61–1.63	.987
Treatment after etoposide
Chemotherapy	1			1
Best supportive care	0.87	0.51–1.47	.594	1.38	0.82–2.31	.220

BMI: body mass index; NACT: neoadjuvant chemotherapy.

When performing multivariable analysis, the FIGO stage IV was found to be independent prognostic factor for PFS and OS, with HR of 3.96 (95%CI: 1.40, 11.17, p = .009) and 3.57 (95%CI: 1.14, 11.20, p = .029), respectively. The main toxicity was grade 2 neutropenia (21.21%). Grade 3 mucositis occurred at 7.58%. Grades III and IV toxicity such as neutropenia, anaemia, mucositis, nausea and vomiting, and thrombocytopenia was 9 (13.64%), 6 (9.09%), 5 (7.58%), 5 (7.58%) and 1 (1.52%), respectively. Dose reduction and delay schedule of chemotherapy were 24 (36.36%) and 11 (16.67%). None of the patients died from a severe side effect.

Discussion

This retrospective study showed favourable effects of oral etoposide. Patients with recurrent or refractory epithelial ovarian cancer or primary peritoneal cancer who had been prescribed oral etoposide at a dose of 50 mg/m² showed the PFS and OS by 3 months and 8 months, respectively. In addition, one patient reported a complete response.

The majority of researches have measured the effectiveness of oral etoposide by the response rate. Oral etoposide has been known for its favourable response rate in ovarian cancer patients from a large phase 2 trial studied by Rose et al. who reported 26.8–34.1% response rate based on 82 participants with a dose of 50 mg/m²/day for 21 days (Rose et al. 1998). Other studies reported the response rate ranging from 6 to 25.8% (Alici et al. 2003; Moosavi et al. 2004; Thavaramara et al. 2009). On the other hand, Kumar et al. used time-based indicators such as PFS and OS to evaluate the effectiveness of oral etoposide (Kumar et al. 2018). Their research included 219 women with high-grade serous epithelial ovarian cancer, the reported PFS, which is from receiving oral etoposide until progression of disease or death, ranged from 2.01 to 2.93 months. Most of their patients were in stage III and received etoposide of 100 mg for 10 days every 3 weeks as a third-line treatment. Their outcome was similar to ours but ours comprised of various histologic subtypes and different dosages of oral etoposide. The study of Kumar et al. also stated that receiving oral etoposide as fourth to eighth line of treatment had a significant longer median OS and PFS than receiving it as earlier line. Conversely, we did not see this significant difference in our study.

Compared with other study conducted in Thailand, Thavaramara et al. reported 4.8 months (range 3.3–6.4 months) of PFS for a daily dose of 75 mg of oral etoposide (Thavaramara et al. 2009). Besides the difference in dosage, other factors can also contribute to such difference. For example, 52% of their patients were prescribed oral etoposide as the second-line treatment, whereas around half of our patients were treated as the third-line chemotherapy. Our study included 13 patients who received one cycle of oral etoposide while Thavaramara et al. excluded them. Moreover, the majority of histopathological subtypes were different; serous carcinoma and clear cell carcinoma accounted for two-thirds of our included patients, but serous carcinoma and endometrioid carcinoma accounted for half of theirs. Also study of Bozkaya et al. with majority of serous papillary carcinoma subtypes has slightly longer OS and PFS (Bozkaya et al. 2017). We hypothesised that clear cell carcinoma which has a worse prognosis played an important role in this response. The comparison of previous retrospective studies to this study is shown in Table 3.

Table 3. Comparison with other previous single agent oral etoposide studies.

Study	N	Dose of oral etoposide per cycle	Median OS (months)	Median PFS (months)
Rose et al. (1998)	99	50 mg/m^2/d for 21 days every 28 days	10.8 (range: 1.9–45.8)	5.7 (range: 0.8–30.8)
Alici et al. (2003)	22	50 mg/m^2/d for 14 days every 28 days	11 (range: 3–36)	2.5 (range: 1–10)
Moosavi et al. (2004)	12	50 mg/d for 21 days every 4 weeks	8.5 (range: 1.2–21.5)	7.5 (range: 5.5–11.0)
Thavaramara et al. (2009)	38	75 mg/d for 21 days every 4 weeks	12.0 (95%CI: 0.75, 25.5)	4.8 (95%CI: 3.3, 6.4)
Bozkaya et al. (2017)	52	50 mg/day for 14 days every 3 weeks	9.95 (95%CI: 0.2, 19.7)	3.2 (95%CI: 2.6, 3.8)
Kumar et al. (2018)	219	100 mg/d for 10 days every 3 weeks	No data	2.01–2.93
The current series	66	50 mg/m^2/d for 21 days every 4 weeks	8.3 (95%CI: 6.8, 10.4)	3.1 (95%CI: 2.3, 3.9)

Haematologic toxicity is the most common adverse effect of oral etoposide. Nine (13.64%) of our patients had grade III or IV neutropenia, while two of phase II studies with similar dosage from Kavanagh et al. and Rose et al. reported 69% and 45.4%, respectively (Kavanagh et al. 1995; Rose et al. 1998). Nevertheless, our study reported 36.36% of dose delay and a 16.67% dose reduction. Despite no death, about one-fourth of our patients discontinued oral etoposide because of its adverse effects.

Although our finding is based upon medical cases with long follow-up periods, limitations of retrospective study are inevitable. Moreover, due to the small sample size in this study, the results from the process of confounder adjustment might be limited and should be cautiously interpretated. The future prospective trials should be performed with more sample size and focussing on other aspects such as assessment of the quality of life. The future research question might be on whether a lower dose will have a comparable outcome and a lower side effect. There are few studies evaluating the role of adding oral etoposide in other drugs. A study of the combination of etoposide with irinotecan has reported a moderate response but with more serious toxicities (Matsumoto et al. 2012). Apatinib combined with oral etoposide in patients with platinum-resistant or platinum-refractory ovarian cancer (AEROC) trial showed a 54% response rate of combined oral etoposide with the oral anti-angiogenic, apatinib (Lan et al. 2018). The effectiveness of combining oral etoposide to other novel targeted therapy such as tyrosine kinase or PARP inhibitor is still in doubt.

In conclusion, administration of oral etoposide at a dose of 50 mg/m² showed fairish oncologic outcomes. It is an option for the palliative aim in platinum resistant patient who refused intravenous chemotherapy. The toxicity profile is also manageable.

Disclosure statement

The authors have no conflict of interest to declare.