A 77-year-old patient was diagnosed with a FIGO IIIC stage high-grade serous ovarian cancer in October 2013. Preoperatively, the patient did not have any preexisting comorbidities and did not take any medication. She underwent a primary complete cytoreduction on 28th October 2013 consisting of en- bloc resection of uterus and adnexa, resection of the rectum with an end- to- end anastomosis, peritonectomy of the pelvis and paracolic gutters, partial resection of the right diaphragm, omentectomy, appendectomy, para- aortic and pelvic lymph node sampling, ureterolysis as well as a subtotal peritonectomy. A tumor of less than 0.5 cm remained in the hilum of the liver. Due to diffuse bleeding, the patient received 1.5 l of colloidal solution, 19 units of fresh frozen plasma and four units of red cell transfusion during the operation. Intraoperatively, a renal ESRD of unknown cause was diagnosed, reducing the glomerular filtration rate from preoperatively > 90 l/min to postoperatively < 15 l/min.

Since that time, the patient required hemodialysis three times a week. Dialysis was performed via an upper arm cephalic vein arteriovenous fistula with a blood flow of 250 ml/min and a dialysate flow of 500 ml/min. A High-Flux dialysis filter (Revaclear 400©) and a Gambro dialysis machine (AK 200©) were used. Unfractionated heparin was administered at a rate of 1000 units/hour after a starting bolus of 1000 IE. The total ultrafiltration rate amounted to 2000 ml with a dialysis session lasting for four hours. Subsequently, an adjuvant chemotherapy with carboplatin 100 mg absolute dose and paclitaxel 175 mg/m2 was initiated in January 2014. However, chemotherapy was discontinued after two cycles as the patient developed delirium with fluctuating cognitive deficiencies, most likely in the context of paraneoplastic syndrome, with spontaneous recovery during the course of time. In August 2015, an embolic stroke of unknown source was diagnosed in the middle cerebral artery and the posterior inferior cerebellar artery leaving the patient with a light ataxia of gait. At the same time, the first platinum-sensitive recurrence was diagnosed with a peritoneal and lymph nodal tumor spread. Due to hemodialysis, she received five cycles of carboplatin monotherapy with a reduced absolute dose of 300 mg per cycle. A partial remission was achieved. In April 2016, maintenance therapy with olaparib capsules with a reduced dose of 200 mg BID was started and taken regularly by the patient for 16 months until progression of disease was diagnosed.

In October 2016, after informed consent of the patient was obtained, venous blood and dialysate samples were drawn from the patient to determine the plasma concentration of olaparib on one dialysis and one non-dialysis day. On the dialysis day, blood samples were collected prior to, 1, 1.5, 2, 3, 4, 5.5, 6 and 8 h after olaparib intake. Hemodialysis started 1.5 h after olaparib intake and lasted for four hours. In doing so, dialysate samples were taken at 1.5, 2.5, 3.5, 4.5 and 5.5 h after olaparib intake. On the non-dialysis day samples were taken prior to, 1, 1.5, 2, 3, 4, 6, and 8 h after olaparib intake. Blood samples were centrifuged for 10 min and stored at − 20 °C.

Plasma concentrations of olaparib were determined using a validated reversed-phase, high-performance liquid chromatography method with TurboIonSpray® tandem mass spectrometric detection in positive ion mode (HPLC–MS/MS). Following the addition of deuterated internal standard ([2H8] olaparib), plasma samples (100 µL) were diluted with water and subjected to manual solid phase extraction on Phenomenex StrataTM-X cartridges. Following elution with acetonitrile and evaporation to dryness, the extracts were reconstituted with HPLC mobile phase (500 µL) and chromatographed on a Waters Xterra® Phenyl 3.5 µm analytical column (50 × 2.1 mm I.D.) eluted at 0.2 mL/min with 1 mM pH3 ammonium formate buffer/acetonitrile (73/27 v/v). Olaparib and internal standard were detected by TurboIonSpray (positive mode) mass spectrometric detection, monitoring ions 435.2 → 281.1 and 443.0 → 281.1 respectively. Calibration curves for Olaparib, prepared in human plasma, were analyzed alongside each batch of samples and the data analyzed using linear regression employing 1/x weighting. The limit of quantification of the assay was 0.5 ng/mL with linearity established over two calibration ranges (0.5–500 ng/mL and 0.02–20 µg/mL). The performance of the assay was monitored throughout use by the inclusion of quality control samples in all bioanalytical runs performed. Assay precision and bias were shown to be within acceptable limits (< 20% at the lower limit of quantification and < 15% at all other concentrations) during study sample analysis.

Dialysate samples were assayed alongside plasma samples using the low range plasma method described previously with the addition of quality control samples prepared in phosphate buffered saline as a non-proteinous surrogate matrix for dialysate. Both the precision and bias of these quality control samples were again shown to be within acceptable limits during study dialysate sample analysis.

Plasma concentration–time data was analyzed using Phoenix™-WinNonlin® v6.3 via non-compartmental analysis. Actual scheduled sample times were used in the pharmacokinetics (PK) analysis to determine multiple dose PK parameters at steady state, namely time to reach maximum plasma concentration (tmax,ss), maximal plasma concentration (Cmax,ss), steady-state trough concentration (Cmin,ss), area under the concentration- time curve (AUC) over 12 h of drug administration (AUCss), AUC until the last measurable concentration after eight hours (AUClast,ss) and apparent plasma clearance following oral administration of the drug (CLss/F). Since samples for bioanalysis were collected up to only 8 h after dosage, concentrations at 12 h after drug administration were estimated based on the terminal elimination rate constants. They were determined based on at least 4 data points.