Roger A. Potish

Erythropoietin increases hemoglobin during radiation therapy for cervical cancer

PURPOSE Anemia during radiation therapy independently predicts poor outcome in patients with cervical cancer. Despite a randomized trial demonstrating red cell transfusions improve local control and survival, many patients are not transfused due to toxicity concerns. This study evaluates the efficacy of recombinant human erythropoietin (r-HuEPO) in reversing anemia in patients undergoing radiation therapy. METHODS AND MATERIALS Twenty patients with criteria of anemia (Hgb < 12.5 g/dL) and surgically staged cervical cancer FIGO stages IB (n = 7), IIA (n = 1), IIB (n = 9), and IIIB (n = 3), ranging in ages from 23-75 years (median 43), were included in this Phase I/II study. Fifteen were treated with r-HuEPO (200 U/kg/day) and ferrous sulfate 5-10 days prior to initiation of external beam radiation therapy, continuing until Hgb was < or = 14 g/dL or completion of radiation therapy. Five patients were treated with ferrous sulfate alone. An additional 61 historical controls meeting eligibility criteria were analyzed. All received external beam radiation therapy and two intracavitary cesium applications. Cisplatinum chemotherapy (20 mg/m2/week) was given as a radiosensitizer in 14 r-HuEPO patients, 4 concurrent controls, and 17 historical controls. RESULTS A marked reticulocytosis was seen in the r-HuEPO group, but not the study controls. In the r-HuEPO group, the mean +/- SD serum Hgb rose + 30% over the course of radiation therapy from a baseline of 10.3 +/- 1.04 g/dL to 13.2 +/- 1.7 g/dL. Average increase in Hgb was 0.5 g/dL per week. Average Hgb during RT was 13.4 g/dL. In study and historical controls, mean initial Hgb levels were 10.7 +/- 1.04 g/dL and 11.1 +/- 1.3g/dL, respectively, remaining unchanged over the course of radiation therapy. Average Hgb levels during radiation therapy were 11.1 g/dL in study controls and 11.4 g/dL in historical controls, significantly lower than r-HuEPO patients (p = 0.0001). Erythropoietin was well tolerated. There were no significant differences in white blood counts (p = 0.6) or platelet counts (p = 0.4) between r-HuEPO patients and both control groups. No patients had blood pressure changes during r-HuEPO therapy. The only possible side effect was deep venous thrombosis, occurring in two patients who were withdrawn from r-HuEPO therapy. Two additional patients developed deep venous thrombosis 9 and 10 days after radiation therapy and r-HuEPO were completed. CONCLUSION Erythropoietin appears to be both safe and effective at raising Hgb levels in anemic cervical cancer patients receiving radiation therapy and chemotherapy.

The role of surgical debulking in cancer of the uterine cervix

From 1978 to 1985, 159 women with advanced cervical cancer received definitive radiation therapy following extraperitoneal surgical staging including pelvic lymph node dissection and periaortic lymph node sampling. Relapse-free survival was a strong function of peritoneal and nodal metastases but was independent of clinical stage. The 5-year relapse-free rate fell from 86% in women without pelvic node metastases to 0% in those with unresectable pelvic node metastases. Women with microscopically positive pelvic node metastases had virtually the same (56%) relapse-free rate as those with grossly positive but completely resected metastases (57%). The overall pelvic failure rate was 16.4% and was significantly higher in women with unresectable pelvic node metastases. Periaortic and peritoneal metastases substantially increased the probability of recurrence. Although histologic grade was prognostically significant, histopathologic category was not. Severe enteric morbidity occurred in 3.6% of patients treated solely to the pelvis and in 7.9% of patients treated to the pelvis and periaortics. Therapeutic implications of debulking pelvic node metastases are discussed.

Enteric morbidity of postoperative pelvic external beam and brachytherapy for uterine cancer

From 1970 through 1986, 219 women received postoperative pelvic external beam therapy and vaginal radium or cesium for uterine cancer. Prescribed external beam and total vaginal surface doses ranged from 38 to 51 Gy and from 70 to 119 Gy, respectively. Severe enteric morbidity developed in 7.8% of patients (15 small bowel, 1 sigmoid, 1 rectal). Complications occurred more frequently in patients with pelvic lymph node sampling at hysterectomy, abdominal surgery prior to hysterectomy, and low body weight. Logistic analysis demonstrated no significant effect of pelvic external beam dose, vaginal surface boost dose, or milligram hours. Five and 10-year overall survival rates were 85% and 74%, respectively. There were two proximal and one distal vaginal recurrences. Recommendations for avoiding complications are presented.

Role of whole abdominal radiation therapy in the management of endometrial cancer; prognostic importance of factors indicating peritoneal metastases

From 1973 through 1983, 27 women received postoperative open-field external beam abdominal radiotherapy as primary treatment of endometrial carcinoma. The 5-year survival rate was 71%. Two distinct prognostic groups were demonstrated. Patients with spread to the adnexa, peritoneal fluid, or both, had a 5-year relapse-free rate of 90%. Patients with macroscopic spread of cancer beyond the adnexa had a 5-year relapse-free rate of 0%. Guidelines are suggested for the radiotherapeutic management of endometrial cancer metastatic to the peritoneal cavity.

Importance of predisposing factors in the development of enteric damage.

The importance of risk factors in the development of radiation-related enteric damage has been demonstrated. Ninety-two women with ovarian tumors received 2000 rads to the abdomen, followed by an additional 2975 rads to the pelvis. By analyzing the development of small bowel obstruction in these women and in other series from the literature, it appears that more than half of enteric complications can be attributed to factors such as prior surgery, thin physique, and preexisting vascular damage. These constitutional factors often will have a greater effect on the development of small bowel damage than will radiation dose, volume, or time. By taking such factors into account and by substituting alternative forms of therapy when feasible, morbidity secondary to radiotherapy can be substantially reduced.

The incidence of second neoplasms following megavoltage radiation for pediatric tumors.

From 1953 to 1975, 330 children received megavoltage for benign or malignant tumors. Fourteen subsequently developed second neoplasms. The 30‐year cumulative incidence of second neoplasms was 9.6%. The incidence of second malignant neoplasms may be lower following megavoltage radiation than following orthovoltage radiation. Cancer 56: 1534‐1537, 1985.

The morbidity and utility of periaortic radiotherapy in cervical carcinoma

From 1971 through 1981, 81 women received 4350 to 5075 rad to the periaortic lymph nodes as part of their primary management for carcinoma of the uterine cervix. While two patients developed chronic small bowel damage, only one required surgical intervention. Five-year disease-free survival was 40%. Approximately one-third of the first recurrences were within the pelvic and periaortic radiation portals, with the remainder in the lungs, liver, bones, abdomen, and supraclavicular lymph nodes. Radiation dose and volume guidelines are presented in order to minimize enteric morbidity.

On the apparent failure of adjuvant pelvic radiotherapy to improve survival for women with uterine sarcomas confined to the uterus.

Despite numerous studies documenting reduction of pelvic relapses after adjuvant pelvic radiotherapy stage I and II uterine sarcomas, improved survival remains unproven. This retrospective report analyzes patterns of failure, survival, and toxicity in 42 women with stage I and 7 patients with stage II uterine sarcomas treated from 1972 through 1998 to identify patients likely to benefit from pelvic or abdominal radiotherapy and chemotherapy. Four of these patients also received adjuvant chemotherapy. There were 20 leiomyosarcomas, 18 homologous mixed mullerian tumors, and 11 heterologous mixed mullerian tumors. Disease-free survivals for mixed mullerian tumors were 65% at 5 years and 61% at 15 years. Disease-free survivals for leiomyosarcomas were 40% at 5 years and 40% at 15 years. There were 14 distant only, 5 distant and abdominal, 1 abdominal, 1 distant and pelvic, and 2 unknown initial sites of failure. Acute toxicity was acceptable as measured by a median 1-kg weight loss from radiotherapy and a 2% rate of failure to complete therapy. Chronic toxicity consisted of 3 small bowel obstructions and 1 sigmoid colon obstruction. In conclusion, the efficacy of adjuvant pelvic radiation is demonstrated by the absence of any isolated pelvic failures. Although the frequent occurrence of peritoneal failures suggests a role for prophylactic abdominal radiation for mixed mullerian tumors, more effective systemic therapy is necessary to substantially increase the chance of cure for women with early-stage uterine sarcomas.

Therapeutic implications of the natural history of advanced cervical cancer as defined by pretreatment surgical staging

From 1978 to 1983, 112 women with advanced cervical carcinomas received radiotherapy after pretreatment surgical staging. Five‐year actuarial relapse‐free survival rates were a strong function of lymphatic spread: 40% with periaortic node metastases, 50% with pelvic node metastases, and 84% without node metastases. Primary treatment failure had a distant component in 75% of recurrences (50% of recurrences with negative nodes and 85% of recurrences with positive nodes). It was concluded that adjuvant systemic therapy is necessary to substantially raise the probability of cure.

Perioperative morbidity and mortality of gynecologic brachytherapy

The incidence and nature of perioperative complications associated with gynecologic brachytherapy applications were determined in 462 implants performed on 327 patients with cervical, endometrial, or vaginal cancer from 1986 to 1988. Twenty‐one patients (6.4%) had life‐threatening complications, including cardiac (16 patients), thromboembolic (4 patients), and pulmonary (1 patient) events. Three complications (0.9%) were fatal within the first 30 days, and two patients died of delayed sequelae of perioperative complications; thus, there was an overall 1.5% mortality rate. Patients experiencing life‐threatening complications tended to be older (P = 0.0002), and to have a history of cardiac disease (P = 0.0002) and an increased American Society of Anesthesiology (ASA) score (P = 0.003). Logistic analysis demonstrated that a history of cardiac disease and increased age were independent predictors of life‐threatening complications (P = 0.01). Careful patient selection and intensive perioperative care of the high‐risk patient may prove to substantially diminish these complications.