ESTRO 38 Abstract book

S819 ESTRO 38

with time constant T.rapid [month] or T.slow [month], and initial quantity Amp.rapid [ng/ml] or Amp.slow [ng/ml]. Mathematical expression of PSA at time t month is as this; PSA(t) = Amp.rapid * exp[- t / T.rapid] + Amp.slow * exp[- t / T.slow] In the assumption that time constant, T.rapid and T.slow might be equal among the all patients, we can also estimate amplitudes of both falls (Amp.rapid and Amp.slow) of each patient using least square method comparing observed and calculated PSA. Results Observed and calculated PSA values were well correlated (R = 0.92). Residuals were almost normally distributed with mean value 0.04 ng/ml and standard deviation 0.29 ng/ml excluding 5 outliers.

Purpose or Objective Significant improvements in the precision of radiotherapy delivery have been made using radio-opaque intra- prostatic markers (IPM). Despite widespread clinical implementation, there is little information available on the frequency and severity of adverse events associated with IPM insertion. The purpose of this study was to quantify patient-reported adverse events following ultrasound-guided, trans-rectal implantation of IPM and to determine the factors that may influence the frequency and severity of those adverse events. Material and Methods A cohort of 100 prostate cancer patients was prospectively accrued. Procedural parameters that may be associated with the risk of adverse events were documented in a data collection form by the interventional radiologist immediately after the IPM insertion. All patients completed an adverse events questionnaire immediately after IPM insertion and approximately 3 to 5 days later. Results All patients were successfully implanted with 3 IPM following a course of prophylactic antibiotics. Local anaesthetic was used in only 2% of patients, and the radiologist reported that only 4% of patients complained of pain. Conversely, 49% of patients reported that the procedure was as painful as, or worse than, previous biopsies. There was no significant change in AUA prostate symptom score after IPM insertion. 23% to 30% of patients reported hematochezia, hematuria or hematospermia over a few days. Fever was reported by 8% of the patients, requiring a visit to their family doctor (6%) or the emergency department (2%) after which they were admitted to hospital for parenteral antibiotic therapy for Cipro-resistant infection. There were no detectable associations between pain during IPM insertion and pain later on, or between use of anticoagulants and bleeding events. Conclusion IPM are a valuable tool for daily image-guidance during prostate radiotherapy but their utility must be balanced with the clinical safety of their implantation. The trans- rectal ultrasound-guided IPM insertion was well tolerated by most patients. Any bleeding or pain experienced was self-limiting. IPM insertion was however, associated with infrequent, but severe infections. EP-1514 Binary exponential model for the PSA fall after IMRT, dependency on initial PSA and Prostate volume H. Nagano 1 , H. Yokoyama 1 , M. Kato 2 , H. Hashimoto 1 , T. Shimo 2 , M. Watanabe 1 , M. Nakanishi 1 , Y. Kaneko 1 , H. Suzuki 1 , A. Noguchi 1 , K. Kobayashi 1 1 Shonan Fujisawa Tokushukai Hospital, Radiation Oncology Department, Fujisawa, Japan ; 2 Tokyo-West tokushukai Hp., Radiology, Akishima, Japan Purpose or Objective PSA (prostate-specific antigen) kinetics for patients with localized prostate cancer after radiation therapy has been reported by some authors. These are mainly analysis of nadir or slops of a certain period. We intended to fit data to binary exponential model mathematically to give away whole course kinetics of PSA after IMRT or VMAT. Material and Methods Twenty one (T1c; 7, T2a; 10, T2b; 2, T2c; 2) patients with mean age 72.4 y. o. (range, 57 - 85 y .o.) and mean GS 6.5 (range, 5 - 8) were treated by IMRT or VMAT (76 Gy / 38 fractions / 8 weeks) without androgen ablation. Their prostate volume before treatment were ranged 13.0 to 68.7 ml with mean value of 39.0 ml. They had mean initial PSA (iPSA) as 7.28 ng/ml (range, 3.58 - 11.85 ng/ml) and had follow-up PSA examinations 2.5 to 8.2 years (mean 4.1 years) with about 3 months interval. We supposed that PSA fall was composed of two steps, rapid and slow ones. Each fall may fit to exponential decay

T.rapid and T.slow were 0.357 and 3.99 months, respectively. Amp.rapid ranged 0.31 to 6.46 ng/ml and mean was 2.93 ng/ml. Amp.rapid correlated well (R = 0.71) with iPSA - Amp.slow (p <0.0005). Amp.slow ranged 0.26 - 1.42 ng/ml and mean was 1.42 ng/ml. Amp.slow correlated somewhat well (R = 0.43) with prostate volume (p < 0.05).

Conclusion PSA fall after IMRT or VMAT was well fitted to our binary exponential model. Amplitude of rapid fall (Amp.rapid) was related to iPSA - Amp.slow, and that of slow fall (Amp.slow) was relevant to prostate volume. We might speculate that slow fall may mainly related to normal prostate recovery after treatment, and rapid fall may represent elimination of cancer cells by radiation therapy.