Practice patterns of intraperitoneal chemotherapy in women with ovarian cancer

ObjectivesTo determine the current patterns of IP chemotherapy use in women with ovarian cancer.MethodsA survey concerning the use of intraperitoneal (IP) chemotherapy was sent to all members of the SGO listed in the 2005 SGO directory and to 200 members of ASCO randomly selected from the ASCO directory in two mailings.ResultsThere were 209 completed responses to the survey, which included 24% of the SGO and 3% of the ASCO members sampled (P   < 0.0001). Of the respondents, 77% indicated that they currently offer IP chemotherapy, 14% either refer patients, or plan to offer IP chemo in the near future, and 9% do not offer IP chemotherapy. The top reasons given by those not using IP chemo was toxicity (63%) and not having the facilities to give an IP infusion (16%). Significant concern over toxicity by all respondents was highest for neurological toxicity (55%) followed by nausea and vomiting (51%), catheter infections (39%) and nephropathy (39%). The most common dose of IP cisplatin was 75 mg/m2 (54%), followed by 100 mg/m2 (38%). Only 6% use carboplatin primarily, but 60% reduce the cisplatin dose based on age. Currently 58% have modified the paclitaxel infusion to allow administration of this regimen in the outpatient setting and 17% plan to do so in the near future. Those using lower doses of platinum are also more likely to offer chemotherapy in the outpatient setting. Concern over toxicity, lack of coverage by residents or fellows, or type of practice did not predict a reduction in platinum dose. A single lumen IP catheter was used by 60% of respondents and a fenestrated is preferred by 40%. This catheter is placed at the time of primary surgery by 58% and 37% routinely place the catheter during a second procedure.ConclusionThe toxicity of IP therapy remains a concern and there does not appear to be any standard IP regimen that is used in women with ovarian cancer. The majority of the SGO members give IP chemotherapy in an outpatient setting at a dose lower than was used in clinical trials. This survey probably represents an overestimate in the use of IP therapy as there is likely a bias in completing the survey by those who currently use this modality. Based on this assumption, it appears that the use of IP chemotherapy by medical oncologists is low.