Prostate Cancer and Prostatic Diseases (2005) 8, 364–368 & 2005 Nature Publishing Group All rights reserved 1365-7852/05 $30.00 www.nature.com/pcan Phase II study of oral capecitabine in patients with hormone-refractory prostate cancer J Spicer1,2, T Plunkett1, N Somaiah1, S Chan1, A Kendall1, N Bolunwu1 & H Pandha1* 1Division of Oncology, Department of Cellular & Molecular Medicine, St George’s Hospital Medical School, Cranmer Terrace, London, UK Background: Currently available treatment for hormone refractory prostate cancer is limited in efficacy and associated with significant toxicity. This phase II study was performed to assess the efficacy of the oral fluoropyrimidine capecitabine in advanced prostate cancer. Patients and methods: Patients who had a rising prostate-specific antigen (PSA) despite androgen withdrawal, but who remained free from cancer-related symptoms. In total, 14 patients received oral capecitabine 1250 mg/m2 twice daily for two weeks of a three-week cycle. Tumour response was assessed using serum PSA measurement at 3-weekly intervals and, where present, imaging of soft tissue metastases. Results: One of 14 patients experienced a partial response as assessed by both PSA and imaging of liver metastases. In seven other patients (50%), treatment decreased the rate of PSA rise. The duration of PSA stabilisation was generally short, but in 5/14 patients (36%) was sustained beyond 18 weeks, and in one patient to 24 weeks. Toxicity was significant but manageable, the most common adverse events being nausea, mucositis and hand–foot syndrome, each occurring in 50% of patients. Other common side effects were diarrhoea and lymphopenia. All toxicities were grade 1 or 2, except for grade 3 hand–foot syndrome occurring in one patient, and no dose reduction was required because of toxicity. Conclusion: Capecitabine has limited activity as a single agent in prostate cancer, but appears to modulate tumour biology. Considering the added convenience of oral administration, these results support further evaluation of combinations containing capecitabine in hormone-refractory prostate cancer. Prostate Cancer and Prostatic Diseases (2005) 8, 364–368. doi:10.1038/sj.pcan.4500821; published online 2 August 2005 Keywords: capecitabine; chemotherapy; PSA Background cer,1,2 and offers some survival benefit.3,4 However, cancer patients express a strong preference for oral rather Patients treated with androgen withdrawal for advanced than intravenous therapies.5,6 Short-lived prostate-speci- prostate cancer become refractory to this therapy within fic antigen (PSA) responses are seen with prednisolone, a median 18–24 months. Combination chemotherapy can estramustine and diethylstilboestrol, but these last two improve symptoms in hormone-refractory prostate can- drugs are associated with significant thrombo-embolic toxicity.7 Most patients remain asymptomatic for some months following detection of PSA rise after androgen *Correspondence: H Pandha, Division of Oncology, Department of withdrawal, so there is a clear need for a well tolerated, Cellular & Molecular Medicine, St George’s Hospital Medical School, orally administered, disease-modifying therapy. Cranmer Terrace, London, SW170RE, UK. The fluoropyrimidine 5-fluorouracil (5FU), adminis- E-mail address: [email protected] 2 tered intravenously with or without folinic acid, pro- Current address: Department of Medical Oncology, Academic duces few responses in hormone-refractory prostate Oncology Offices, 3rd Floor, Thomas Guy House, Guy’s Hospital, St cancer, at the expense of significant toxicity.8–12 However, Thomas Street, London SE1 9RT, UK 13,14 Received 23 March 2005; revised 19 June 2005; accepted 19 June 2005; responses were seen using low-dose infusional 5FU. published online 2 August 2005 Capecitabine (Xeloda, Roche Products) is an orally Study of oral capecitabine J Spicer et al administered fluoropyrimidine carbamate preferentially described.23 Androgen withdrawal with an LHRH 365 converted to 5-FU at sites of disease by thymidine analogue was continued in all patients. Treatment was phosphorylase expressed in tumour cells,15 including given if the neutrophil count exceeded 0.5 Â 109/l, prostate cancer cells.16 Capecitabine is active in a number lymphocyte count exceeded 0.5 Â 109/l and platelets of tumour types including colorectal, breast, stomach were greater than 100 Â 109/l. Adverse events were and pancreatic cancer. In particular, capecitabine mono- graded on a four-point scale (WHO common toxicity therapy is now well established as first-line treatment in criteria), with an additional scale for hand–foot syn- advanced colorectal cancer,17,18 and as a preferred option drome. Capecitabine was delayed and dosage reduced in taxane-pretreated metastatic breast cancer.19 Survival by 20% for treatment-related adverse events of grade 2 or with the combination of capecitabine and docetaxel is greater. Patients were eligible to receive eight cycles, with superior to docetaxel alone in metastatic breast cancer.20 the option to continue in the event of maintained There is anecdotal evidence of capecitabine activity in response. Treatment was discontinued in the absence of prostate cancer,21 and we have conducted an open label PSA response or control (a fall in PSA velocity by 50% or phase II study to evaluate the efficacy and safety of this more) after four cycles, on PSA progression subse- drug in patients with hormone-refractory prostate quently, or on symptomatic deterioration. cancer. PSA response was chosen as the primary end point, according to published guidelines.22 Patient assessment and response criteria Patients and methods Screening assessments, including a medical history, physical examination, full blood count and chemistry Study design profile were performed within 1 week before treatment began. Serial rises in serum PSA, as above, were required This open label, phase II, non-randomised single-centre before trial entry. PSA level and full blood count were study was designed to investigate the efficacy of repeated with each 21-day cycle. A PSA response was capecitabine in prostate cancer patients with progressive defined as a X50% decrease from baseline in serum PSA, disease despite androgen withdrawal. The study re- determined by two observations not less than 4 weeks ceived ethical approval from the local ethics review apart. PSA velocity was defined as the rate of change of committee, and all patients gave written, informed the natural logarithm of PSA with time, and baseline PSA consent. velocity was calculated using at least two pre-treatment PSA values for each patient. PSA stabilisation was defined as a fall in PSA velocity by 50% or more. Patients Patient eligibility meeting neither of these criteria were considered to have progressive disease. For patients with measurable dis- Eligible patients had histologically confirmed adenocar- ease, radiological investigations were repeated after cinoma of the prostate with progression of locally three cycles, or at biochemical or clinical progression. advanced or metastatic disease despite androgen with- Response was assessed according to RECIST criteria,24 drawal with a luteinising hormone releasing hormone with partial response defined as a X30% decrease from (LHRH) agonist. Disease progression was defined as a baseline. rise in serum PSA on two consecutive occasions at least 2 weeks apart.22 Patients with measurable disease were also included. Statistical methods Patients who had received prior chemotherapy were not eligible. Previous or concurrent radiotherapy was The aims of this phase II study were to assess the efficacy allowed. Patients were required to have a WHO and toxicity of capecitabine in hormone refractory performance status of 2 or less, and to have a life prostate cancer. The primary end point was response as expectancy of at least 3 months. Those with significant assessed using serum PSA. Using a sequential analysis weight loss or severe pain were excluded. Patients with design,25 the absence of a response among the first 14 inadequate haematological, renal or liver parameters patients would indicate insufficient study drug activity (neutrophils o0.5 Â 109/l, calculated creatinine clearance to warrant further study. Thus 14 assessable subjects o50 ml/min, platelets o100 Â 109/l, bilirubin 430), were recruited. intracranial metastases or cord compression were also The analysis of efficacy was based on all patients who excluded. Other exclusion criteria included any concur- received at least one dose of capecitabine. Similarly, rent medical condition or laboratory abnormality that safety was assessed for all patients who received at least could compromise the safety of the patient or interfere one dose of the study drug. with the interpretation of the results. Treatment regimen Results Patients were treated with oral capecitabine 1250 mg/m2 A total of 14 patients were enrolled in this study between self-administered twice daily (2500 mg/m2/day), as July 2002 and November 2003. All patients completed at intermittent therapy in 3-week cycles consisting of 14 least one cycle of capecitabine and were assessable for days of treatment followed by 7 days without treatment. response. All patients had progressed following LHRH Details of the treatment regimen have been previously therapy. The median PSA at study entry was 496 ng/ml, Prostate Cancer and Prostatic Diseases Study of oral capecitabine J Spicer et al 366 Table 1 Baseline characteristics of patients treated 3 Age Performance Sites of Baseline PSA 2 status metastatic (ng/ml) disease 1 1 72 1 Bone 724 2 73 1 Bone 15 0 3 58 0 Bone 407 1 2 3 4 5 6 7 8 9 10 11 12 13 14 4 63 0 Bone 304 -1 5 56 1 Bone; pelvic LNa 40 PSA velocity ratio PSA velocity 6 77 1 Bone 967 -2 7 73 2 Bone 176 8 41 1 None 1308 -3 9 74 1 Bone 100 Patient number 10 70 1 Liver 1474 11 69 1 Bone & liver 796 Figure 2 PSA velocity change. PSA velocity is defined as the ratio of rate 12 79 1 Bone 107 of change in ln (post-treatment PSA) to the rate of change in ln (pre- 13 69 0 Bone 217 treatment PSA).