Ruth Peter

Late GI and GU complications in the treatment of prostate cancer

PURPOSE To assess the factors that predict late GI and GU morbidity in radiation treatment of the prostate. METHODS AND MATERIALS Seven hundred twelve consecutive prostate cancer patients treated at this institution between 1986 and 1994 (inclusive) with conformal or conventional techniques were included in the analysis. Patients had at least 3 months follow-up and received at least 65 Gy. Late GI Grade 3 morbidity was rectal bleeding (requiring three or more procedures) or proctitis. Late Grade 3 GU morbidity was cystitis or stricture. Multivariate analysis (MVA) was used to assess factors related to the complication-free survival. The factors assessed were age, occurrence of side effects > or = Grade 2 during treatment, irradiated volume parameters (use of pelvic fields, treatment of seminal vesicles to full dose or 57 Gy, and use of additional rectal shielding), dose, comorbidities, and other treatments (hormonal manipulation, TURP). RESULTS Acute GI and GU side effects (Grade 2 or higher) were noted in 246 and 201 patients, respectively; 67 of these patients exhibited both. GI side effects were not correlated with GU side effects acutely. Late and acute morbidities were correlated (both GI and GU). Fifteen of the 712 patients expressed Grade 3 or 4 GI injuries 3 to 32 months after the end of treatment, with a mean of 14.3 months. One hundred fifteen patients expressed Grade 2 or higher GI morbidity (mean: 13.7 months). The 43 Grade 2 or higher GU morbidities occurred significantly later (mean: 22.7 months). Central axis dose was the only independent variable significantly related to the incidence of late GI morbidity on MVA. No treatment volume parameters were significant for Grade 3. The following parameters were significantly related (by MVA) to Grade 2 GI morbidity: central axis dose, use of the increased rectal shielding, androgen deprivation therapy starting before RT. Acute and late GI morbidities were highly correlated. History of diabetes, treatment of pelvic nodes, and age less than 60 years were significantly related to acute GI side effects. The parameters significantly related to late Grade 2 or higher GU morbidity were central axis dose, androgen deprivation therapy (Zoladex or Lupron) prior to radiation therapy (RT), history of obstructive symptoms, and acute GU side effects. There were too few late Grade 3 GU morbidities to perform multivariate analysis. Acute GU side effects were highly correlated with late GU injury. The following were correlated with acute GU side effects: history of diabetes (+), treatment with conformal fields (-), TURP before RT (-), presentation with urinary obstructive symptoms. CONCLUSION Both late GI and GU morbidity demonstrate a dose dependence, but only the volume dependence observed is a reduction in late Grade 2-4 GI morbidity by increasing the rectal shielding in the lateral fields for the final 10 Gy. Moreover, both late GI and GU morbidity was increased in patients treated with hormone manipulation prior to RT. GI and GU injuries were correlated with their corresponding acute side effects. GI and GU complications must not be combined for analysis to determine the factors related to their occurrence.

Chronic rectal bleeding after high-dose conformal treatment of prostate cancer warrants modification of existing morbidity scales.

PURPOSE Serious late morbidity (Grade 3/4) from the conformal treatment of prostate cancer has been reported in <1% to 6% of patients based on existing late gastrointestinal (GI) morbidity scales. None of the existing morbidity scales include our most frequently observed late GI complication, which is chronic rectal bleeding requiring multiple fulgerations. This communication documents the frequency of rectal bleeding requiring multiple fulgerations and illustrates the variation in reported late serious GI complication rates by the selection of morbidity scale. METHODS AND MATERIALS Between May 1989 and December 1993, 352 patients with T1-T3 nonmetastatic prostate cancers were treated with our four-field conformal technique without special rectal blocking. This technique includes a 1-cm margin from the clinical target volume (CTV) to the planning target volume (PTV) in all directions. The median follow-up for these patients was 36 months (range 2-76), and the median center of prostate dose was 74 Gy (range 63-81). Three morbidity scales are assessed: the Radiation Therapy Oncology Group (RTOG), the Late Effects Normal Tissue Task Force (LENT), and our modification of the LENT (FC-LENT). This modification registers chronic rectal bleeding requiring at least one blood transfusion and/or more than two coagulations as a Grade 3 event. Estimates for Grade 3/4 late GI complication rates were determined using Kaplan-Meier methodology. The duration of severe symptoms with chronic rectal bleeding is measured from the first to the last transrectal coagulation. Latency is measured from the end of radiotherapy to surgery, first blood transfusion, or third coagulation procedure. RESULTS Sixteen patients developed Grade 3/4 complications by one of the three morbidity scales. Two patients required surgery (colostomy or sigmoid resection), three required multiple blood transfusions, two required one or two blood transfusions, and nine required at least three coagulations. The median duration of bleeding for those patients requiring multiple procedures was 7 months (range 3-33) and the median latency was 22 months (range 9-40). The 5-year actuarial rate of Grade 3/4 complications by each scale are: RTOG 0.7%, LENT 2%, and FC-LENT 6%. The rate of chronic rectal bleeding increases with increasing dose and is low in patients treated with conventional techniques owing to lower doses. CONCLUSION Chronic rectal bleeding requiring any blood transfusion(s) or multiple coagulation procedures is our most frequently observed complication. This complication appears late in follow-up and is present for a long duration. We believe this justifies the inclusion of chronic rectal bleeding requiring multiple coagulation procedures as a Grade 3 event in future morbidity scales. Our data illustrate that published Grade 3/4 morbidity rates are highly dependent on the morbidity scale selected, as our data show 0.7% RTOG, 2% LENT, and 6% FC-LENT. Obviously, a uniform scale is required that includes the newly recognized serious late effects associated with the conformal treatment of prostate cancer.

Rectal bleeding after conformal 3D treatment of prostate cancer: Time to occurrence, response to treatment and duration of morbidity

PURPOSE Rectal bleeding is the most common late sequelae of high-dose 3D conformal treatment (3DCRT) for prostate cancer and may limit attempts to improve local control by dose escalation. The clinical course of this complication is reported including time to onset, response to treatment, duration of morbidity, and multivariate analysis for predictors. METHODS AND MATERIALS From March 1989 to June 1995, 670 patients with prostate cancer were treated with 3DCRT at Fox Chase Cancer Center. Eighty-nine patients developed Grade 2 or Grade 3 complications due to rectal bleeding and are analyzed. Multivariate analysis results for predictors of Grade 2 and 3 rectal bleeding are reported as well as time to development, response to initial and retreatment, and duration of morbidity. RESULTS The median time to occurrence is not significantly different (p = 0.09) for Grade 2 (13 months, range 4-41 months) compared to Grade 3 rectal bleeding (18 months, range 4-40 months), while the corresponding median duration of symptoms was significantly different (p < 0.0001) being 1 month (range 1-12) vs. 10 months (1-34) for Grade 2 and Grade 3 bleeding, respectively. For Grade 2 bleeding, medication or coagulation was highly effective as initial or retreatment resolving 66 of 73 patients. For Grade 3 bleeding, three patients responded without medication following blood transfusion only, while with multiple coagulations and medication 12 of 16 patients improved to < or = Grade 1. Multivariate analysis demonstrates that dose is the only significant factor associated with Grade 2 (p = 0.01) or Grade 3 (p = 0.01) rectal bleeding. Of seven nonresponders to treatment for Grade 2 bleeding, three have died of intercurrent disease at 10, 19, and 26 months, while four are alive with continuing Grade 2 bleeding at 12, 14, 15, and 30 months after onset. The four nonresponders to treatment for Grade 3 bleeding continue to bleed 1, 9, 32, and 35 months after the third coagulation despite continuing care. CONCLUSIONS Chronic rectal bleeding is a sequelae of high-dose conformal treatment of prostate cancer. Grade 2 morbidity responds to medication or limited coagulation (< or = 2) in 90% of patients. Grade 3 morbidity responds to medication and multiple coagulations (> or = 3) in 75% of patients. The chronicity of Grade 3 morbidity is illustrated by a 10-month median duration for response to treatment, with a range of response extending to 34 months. Nonresponders to treatment have continued to bleed up to 35 months after the third coagulation. Appropriate shielding of the rectal mucosa limiting dose to < 72 Gy is required to avoid a high incidence of these complications, as dose is the only significant variable associated with rectal bleeding.

Quality of life study in prostate cancer patients treated with three-dimensional conformal radiation therapy: Comparing late bowel and bladder quality of life symptoms to that of the normal population☆

PURPOSE The goals of this study were twofold. First, differences were quantified for symptoms that impact bowel and bladder quality of life (QOL) in prostate cancer patients treated with three-dimensional conformal radiotherapy (3DCRT) alone to the prostate vs. whole pelvis with prostate boost. Second, bowel and bladder QOL measures for these patients were compared to those of the normal population of men with a similar age distribution. MATERIAL AND METHODS Two health status surveys evaluating bowel and bladder functioning, along with the AUA Symptom Problem Index and the BPH Impact Index, were mailed to 195 prostate cancer patients treated with 3DCRT between 12/92 and 11/95 at Fox Chase Cancer Center by a single clinician (GH). No patient received hormonal management as part of his treatment. Ninety-five patients had pretreatment PSA levels <10 ng/ml, T1/T2A tumors with Gleason scores 2-6, and no perineural invasion. They were treated to the prostate alone and are referred to as Group I. The remaining 100 patients had one or more of the following characteristics: pretreatment PSA levels > or =10 ng/ml, T2B/T3 tumors, Gleason scores 7-10, or perineural invasion. These patients were treated to the whole pelvis followed by a boost to the prostate and are referred to as Group II. Frequencies were tabulated, and differences in percentages for the two groups were evaluated using the two-tailed Fishers Exact Test. Overall percentages were compared to those for equivalent measures reported by Litwin (1999) based on a normal population of men with a mean age of 73 years (range 47-86). Comparisons to the normal population were also evaluated using two-tailed Fishers Exact p values. RESULTS The mailing yielded a high response rate of 71% (n = 139, 66 in Group I and 73 in Group II). The mean age was 67 (range 49-82), and the median ICRU dose levels for Groups I and II were 73 and 76 Gy, respectively. Responses relating to bladder symptoms were similar for Groups I and II, except for the degree of bother associated with trouble in urination over the last month. Percentages for no bother at all were 66% and 56% for Groups I and II, respectively. Observed differences in bowel functioning related to rectal urgency over the past year (22% vs. 40% for Groups I and II, p = 0.03), the use of pads for protection against bowel incontinence (0% vs. 10% for Groups I and II, p = 0.01), and bowel satisfaction (88% vs. 72% for Groups I and II, p = 0.03). There was no significant difference in the degree of bother bladder symptoms cause men treated with radiotherapy as compared to men without cancer. Few patients reported bowel dysfunction bother as a big problem, but patients do tend to have more very small to moderate bother from bowel dysfunction than the normal population (55% vs. 33%, p < 0.001). CONCLUSION This is the first long-term study of QOL in men treated with high-dose 3DCRT for prostate cancer. It demonstrates that these men enjoy QOL related to bladder function similar to that of the normal population. Few patients report bother from bowel symptoms as a big problem but tend to have more very small to moderate bother than the normal population. Treatment of prostate cancer patients to the whole pelvis may result in decreased QOL as defined by rectal urgency, the use of pads for bowel incontinence, and satisfaction with bowel functioning. However, regardless of field size, men are generally satisfied with their bowel and bladder functioning three to six years post treatment.

Phase I trial of 5-fluorouracil, leucovorin, and cisplatin in combination

SummaryThe ongoing evaluation of combination chemotherapy with 5-fluorouracil (5-FU) and cisplatin in several tumors prompted a phase I clinical trial of cisplatin with 5-FU modulated by leucovorin. A total of 26 patients were treated with varying doses of 5-FU by continuous i.v. infusion for 5 days; 200 mg/m2 leucovorin was given by daily bolus injection for 5 days; and 20 mg/m2 cisplatin was infused over 2 h on each day of treatment. Courses were repeated every 21–28 days. The starting dose of 5-FU was 300 mg/m2. Poor-risk patients (extensive prior radiation, performance status of 2 or worse) did not tolerate the initial dose; the maximum tolerated dose of 5-FU in this group was 200 mg/m2 daily. Good-risk patients tolerated 300 mg/m2, but a majority had excessive toxicity at higher doses. The dose-limiting toxicity was gastrointestinal (mucositis/diarrhea) and/or myelosuppression; additional side effects included were nausea and vomiting (≤grade 2) and ataxia (one patient). Among 13 patients with colorectal cancer, 4 partial responses were observed. The marked reduction in the tolerable dose of 5-FU occasioned by the addition of modulating doses of leucovorin is noteworthy. The response observed support further investigation of this regimen in phase II trials.