Octreotide Acetate in Prevention of Chemoradiation-Induced Diarrhea in Anorectal Cancer: Randomized RTOG Trial 0315

Abstract

Background

In anorectal cancer patients, an acute side effect of chemoradiotherapy is gastrointestinal toxicity, which often impedes treatment delivery. Based on previous trials, octreotide acetate is widely recommended for the control of chemotherapy-induced diarrhea. However, the effectiveness of octreotide in preventing or controlling radiation- and chemoradiation-induced diarrhea is not known.

Methods

A randomized, double-blinded, placebo-controlled trial was designed to determine the efficacy of long-acting octreotide acetate (LAO) in preventing the onset of acute diarrhea in patients undergoing chemoradiation therapy for rectal or anal cancer. Between 4 and 7 days before the start of radiation therapy, patients received a 30-mg dose of LAO (109 patients) or placebo (106 patients) via intramuscular injection. A second dose was given on day 22 (±3 days) of radiation treatment. A total of 215 patients were included in the final analysis. The primary endpoint was the incidence of grade 2–4 acute diarrhea; secondary endpoints included treatment compliance, medical resource utilization, patient-reported bowel function, and quality of life (QoL). Statistical tests were one- or two-sided, as specified.

Results

After a median follow-up time of 9.64 months, incidence rates of grades 2–4 acute diarrhea were similar in both groups (49% placebo vs 44% LAO; P = .21). No statistically significant treatment differences in chemotherapy or radiation delivery, medical resource utilization, patient-reported bowel function, or QoL were observed.

Conclusion

In this study, the prophylactic use of LAO did not prevent the incidence or reduce the severity of diarrhea and had no notable impact on patient-reported bowel function or QoL.

CONTEXT AND CAVEATS

Prior knowledge

Octreotide acetate is used to control chemotherapy-induced acute diarrhea in several solid cancers. It is not known whether octreotide acetate can prevent and control acute diarrhea in patients receiving concurrent chemotherapy and radiation treatment (chemoradiation) for anorectal cancer.

Study design

A randomized double-blinded trial was designed to evaluate the efficacy of long-acting octreotide acetate or placebo to reduce the incidence and severity of acute diarrhea in anorectal cancer patients receiving chemoradiation. Secondary endpoints included patient-reported quality of life, bowel function, treatment compliance, number of hospitalizations, or utilization of medical resources.

Contribution

No statistically significant difference was observed between the placebo and long-acting octreotide acetate groups in the incidence of mild, moderate, or severe diarrhea; hospitalizations; quality of life; bowel function; or utilization of medical resources.

Implications

The use of long-acting octreotide acetate for the prevention of chemoradiation-induced diarrhea was ineffective.

Limitations

The study did not account for the surgical status (pre- vs postoperative) of the patients and the bowel-sparing techniques that can reduce the incidence of diarrhea.

From the Editors

Colorectal cancer is a major cause of morbidity and mortality among both men and women in the United States. Approximately 146 970 people were diagnosed with colon and rectal cancers, and approximately 49 920 died of these diseases in 2009 according to the estimates provided by the American Cancer Society (1). The incidence data include approximately 40 870 new cases of rectal cancer (23 580 men and 17 290 women) (1). In contrast approximately 5290 people were diagnosed with anal cancer in 2009 and approximately 710 people died of this disease. In contrast, incidence of anal cancer is greater among women (3190 new cases) than among men (2100 new cases). Similarly, mortality from anal cancer is greater among women (450 deaths) compared with men (260 deaths) (1). For localized rectal cancer treated with either pre- or postoperative chemoradiation therapy, local regional recurrence rates are in the range of 6%–14% and overall survival is in the range of 57%–76% (2–6). The concurrent chemoradiation therapy for localized anal cancer results in complete response rates in the range of 80%–92% (7,8) and 5-year overall survival in the range of 59%–65% (9–11).

Acute gastrointestinal toxicity is often a dose-limiting side effect of chemoradiation for anorectal (rectal and anal) cancer. The overall incidence of grades 3–4 acute diarrhea in stages II and III rectal cancer is 16%–39% in preoperative radiotherapy trials using 5-fluorouracil (5-FU)–based concurrent chemotherapy; it is 20%–30% among patients receiving postoperative chemoradiotherapy (6,12,13). These rates of acute diarrhea become even higher with currently investigated chemoradiation regimens for rectal cancer using capecitabine, oxaliplatin, and pelvic radiation (14). Similarly, acute diarrhea is also a dose-limiting toxic effect with the administration of 5-FU, mitomycin, and pelvic radiation for anal cancer. The rates of grades 2–3 gastrointestinal toxicity are reported to be 31%–38% (15). The associated diarrhea is often severe enough to require dose reduction or discontinuation of chemotherapy and/or delays in radiotherapy. The key to successful treatment is uninterrupted delivery of planned treatment doses with tolerable normal tissue toxicity and without compromising the curative potential.

To date, there is no standard care for the prevention or reduction of chemoradiation-induced diarrhea, which is often characterized by increased frequency of stools, increased fecal urgency, altered consistency of stools, abdominal pain or discomfort, and the inability to distinguish stool from gas (16–18). Other less commonly documented symptoms such as tenesmus, mucus, blood, or dehydration may also be associated with uncontrolled diarrhea. The intensity of these symptoms may vary among individual patients; however, these symptoms may markedly interfere with patients’ daily activities and overall quality of life (QoL). Generally, the diarrhea is treated symptomatically with antidiarrheal agents, such as loperamide, diet modification, and hydration (19–21). Octreotide acetate, a somatostatin analog, is also known to reduce intestinal motility and pancreatic secretions and to promote intestinal absorption (22,23). It is approved by the US Food and Drug Administration for the treatment of diarrhea associated with metastatic carcinoid tumors and vasoactive peptide–secreting tumors (24). Octreotide has also been shown to control grades 3–4 diarrhea induced by chemotherapy for a variety of solid malignancies. In 80%–95% of patients unresponsive to the more commonly used antidiarrheal agents such as loperamide, short-acting octreotide (SAO) is able to control grades 3–4 diarrhea associated with chemotherapy (25–27). Phase III trials have demonstrated the effectiveness of octreotide in controlling 5-FU–induced diarrhea as well as for the secondary prevention of diarrhea in patients receiving cisplatin (25–29). Octreotide is widely recommended for the treatment of chemotherapy-induced diarrhea (19–21).

However, the utility of octreotide for the prevention of radiation- or chemoradiation-induced diarrhea at the inception of this trial was unknown. The primary objective of the Radiation Therapy Oncology Group (RTOG) 0315 was to determine the efficacy of long-acting octreotide (LAO) in preventing the onset of acute diarrhea in patients receiving concurrent chemoradiation for rectal or anal cancer. LAO was chosen over the three-times daily dose of SAO for patient convenience. It is also worth mentioning that in addition to patient care, treatment delays also have financial implications for both the patient and the health-care system. It is crucial to minimize excess use of health-care resources because of delays or interruption in treatment. Thus, the secondary objectives included assessing the impact of LAO on chemoradiation delivery, medical resource utilization, patient-reported bowel function, and QoL.

Patients and Methods

Trial Design

This was a randomized, double-blinded, placebo-controlled trial. A total of 233 patients with rectal and anal cancers were enrolled for this study from 16 RTOG full-member, 41 RTOG affiliate, and 15 RTOG Community Clinical Oncology Program (CCOP) member institutions. The trial design is shown in Figure 1. The trial was registered with ClinicalTrials.gov as NCT00075868.

Figure 1.

CONSORT diagram.

Eligibility Criteria

Patients receiving concurrent chemotherapy and pelvic radiation therapy for rectal or anal cancer were considered eligible for the study. Patients with a history of chemotherapy were also allowed to participate. Patients with prior pelvic irradiation, hyperfractionated radiotherapy, split-course radiotherapy, intensity-modulated radiotherapy, and/or planned brachytherapy before completion of external-beam radiation therapy were excluded from the study. Patients with a known allergy/hypersensitivity to octreotide acetate or other related drugs/compounds and patients previously receiving octreotide acetate for cancer therapy–related diarrhea were also excluded. Patients with a history of chronic or acute regional enteritis, malabsorption syndrome(s), or other history of inflammatory bowel disease; patients with uncontrolled diabetes and cholecystitis or gallstones (unless cholecystectomy was performed); patients with colostomy or those who had abdominoperineal resection or other surgery leaving the patient without a functioning rectum were also excluded. Patients with any incontinence of stool or uncontrolled diarrhea (>grade 2 Common Terminology Criteria for Adverse Events version 3.0 [CTCAE v3.0]) (17) at baseline were excluded. Only 17% and 18% of the patients experienced grades 1–2 diarrhea before study entry, and only 3% of patients in each group had grade 2 diarrhea at study entry. To address the lack of standard patient-reported diarrhea assessment tools (19,20), four patient-reported bowel function measures were also used in conjunction with the clinician-rated CTCAE v3.0 measure.

We excluded patients who received investigational drugs within 30 days of study entry. No glucocorticoid therapy, insulin sensitizers, or exogenous growth hormone therapy were allowed within 6 months of study entry. Pregnant or lactating women, patients with a history of hepatic disease, and HIV-positive patients were also excluded. The institutional review board at each participating institution approved this study. All patients were at least 18 years of age and signed a study-specific informed consent form before participating.

Randomization and Treatment Allocation

Between December 5, 2003, and February 24, 2006, a total of 233 patients were randomly assigned to receive two 30-mg intramuscular injections of LAO or placebo. Fourteen patients did not meet the eligibility criteria or withdrew their consent for participation in the trial after randomization (Figure 1). Although the less expensive 20-mg dose has proven effective for chemotherapy-induced diarrhea (30), the 30-mg dose was chosen to maximize efficacy in this trial (28,31,32). All patients first received a 100-μg test dose of LAO to assess for sensitivity or allergy to the drug before the first dose of study drug administration. Patients who exhibited symptoms of intolerance within 24 hours of the test dose were not given the study medication but received chemoradiation, per protocol. Patients who demonstrated tolerance received the first dose of study drug between 4 and 7 days before the start of radiation because a latent period before the drug was expected to be fully effective. The second dose of study drug was given on day 22 (±3 days) of radiation. Patient demographic and stratification variables were well balanced between the placebo and LAO treatment groups. Eight percent of patients were of Hispanic/Latino descent, 9% of patients were African American, 89% were Caucasian, and fewer than 2% were Native American or Asian.

Randomization was performed using the Zelen (33) treatment allocation scheme to balance patient factors and institution. Institutions received vials of test dose and study drug identified by patient case number only. The majority of patients had rectal cancer (80%) and were expected to receive continuous chemotherapy (82%) and radiation of more than 50 Gy (83%). The actual total radiation dose received is not reported because of inconsistencies in the reporting of boost fields. Patients were stratified by planned radiotherapy dose (<50 Gy and ≥50 Gy), planned chemotherapy dose (bolus or continuous infusion), and sex. Patients received concurrent chemotherapy and radiation therapy to the pelvis. The superior border was at the top of the sacroiliac joint in 67% of the pelvic radiation fields. The chemotherapy regimen was per institutional choice. Patients received a continuous course of definitive or adjuvant external-beam pelvic radiation therapy to a minimum dose of 45 Gy, with a daily fraction size of 1.5–2 Gy administered four to five times per week. The radiotherapy portal was required to be greater than 10 × 10 cm. The superior border of the field was limited by the L4-5 interspace. The total planned dose to the central axis midplane (for anteroposterior/posteroanterior parallel opposed fields) or isocenter (for three- or four-field techniques) for the whole pelvic field was in the range of 45–50.4 Gy, with a boost allowed to the tumor bed. During treatment, patients were evaluated weekly for adverse events by a physician using the CTCAE v3.0 criteria. Patients were also assessed weekly for the intervention of medical regimens to control diarrhea or its complications and the need for any outpatient or hospital-based care for management of enteritis or its complications. Patients were then assessed at 3, 6, 9, and 15 months from start of radiation therapy to monitor treatment effects, diarrhea, and health-care resource utilization.

Patient-Reported Outcomes

Patients completed four validated self-administered QoL and function assessment tools before treatment (baseline), within 2 weeks of completion of radiotherapy, and quarterly for 15 months. The baseline and 3-month assessments were of particular interest to match with the interval of the diarrhea and toxicity analysis. The patient-reported outcomes were the 24-item health-related Quality of Life-Radiation Therapy Instrument (QOL-RTI) (34,35), the 14-item Expanded Prostate Cancer Index Bowel (EPIC-Bowel) (36,37), the seven-item Functional Alterations due to Changes in Elimination Bowel (FACE-Bowel) (38), and the four-item Diarrhea Assessment Scale (DAS) (39). Composite scores of each tool were computed using the guidelines associated with each instrument. Missing item values were imputed using the average values of the preceding and following completed items. For the QOL-RTI and the EPIC-Bowel, higher scores indicate better QoL or bowel function, and for the FACE-Bowel and DAS scales, lower scores indicate better bowel functioning. In this study, if octreotide acetate was efficacious in reducing treatment-induced diarrhea, better QoL and bowel scores were expected for the octreotide acetate group for all instruments. Details of the psychometric properties (reliability and validity) of each instrument are summarized in Supplementary Materials (available online).

Statistical Design and Analysis

The primary endpoint of this trial was the incidence of moderate or severe acute diarrhea (CTCAE v3.0, grades 3–4). Assuming that the incidence of grades 3–4 diarrhea in the placebo group would be 35%, 214 patients would be required to ensure 90% statistical power to detect a 50% decrease in grades 3–4 diarrhea with LAO treatment, which is based on the two-sample binomial test at the α = 0.05 level (one-sided; increase in grades 3–4 diarrhea because of LAO treatment was not anticipated). Two interim analyses were planned using the Lan-DeMets spending function for efficacy testing, occurring when 50% (n = 107) and 75% (n = 161) of patients were accrued and followed for a minimum of 3 months. Nominal statistical significance levels for the interim and final analyses were 0.0056, 0.0219, and 0.0427, respectively (40). Logistic regression models were used to evaluate treatment differences adjusted for important baseline variables.

The secondary endpoints of this trial included treatment compliance, health-care resource utilization, patient-reported bowel function, and QoL. The standard error of measurement (SEM) was calculated based on the reliability coefficients for each patient-reported assessment and the patient scores on the baseline assessment. The reliability coefficients based on the baseline assessments were QOL-RTI, 0.85; EPIC-Bowel, 0.69; FACE-Bowel, 0.78; and DAS, 0.55. Patient change scores were classified as deteriorating, stable, or improved based on the deviation from the SEM. Each patient’s 3-month scores within 1.0 SEM of the pretreatment score were classified as stable; scores at least 1.0 SEM better than the pretreatment score were classified as improved; and scores at least 1.0 SEM worse than the pretreatment score were classified as deteriorated. Assuming that 60% of patients in the placebo group would deteriorate at 3 months on a given assessment, a sample size of 214 patients would ensure at least 87% statistical power to detect a decrease from 60% to 40% in the proportion of patients that deteriorated based on the one-sided χ²-square test at the 0.05 significance level because the LAO treatment was anticipated to improve outcomes on all four assessment tools. In addition to the trial endpoints, the relationships between the patient-assessed EPIC, FACE, and DAS tools and the physician-assessed CTCAE diarrhea tool were evaluated. Assessments at all follow-up times were used in the analysis. Spearman correlation coefficients were computed using the bootstrap method to adjust for multiple assessments per patient (41). All data were analyzed with SAS (version 9.1.3 for Windows; SAS Institute, Cary, NC) and R (version 2.7.2 for Windows; R Foundation for Statistical Computing, Vienna, Austria).

Results

The pretreatment characteristics for all eligible patients are shown in Table 1. The results of the two interim analyses did not warrant early stopping of the trial or early reporting of the results. However, after the first interim analysis, because of lower-than-expected incidence rates of grade 3 and higher diarrhea, the primary endpoint was modified. This modification was achieved without loss of statistical power and with institutional review board approval to include mild diarrhea (grade 2). Two hundred fifteen patients were included in the final analysis of grades 2–4 acute diarrhea. Patients were not excluded because of treatment noncompliance; however, four eligible patients were excluded because they were lost to follow-up and missing adverse event information. The median follow-up time for patients was 9.64 months (range 0.60–22.3 months). We did not observe a statistically significant treatment difference in the primary endpoint. Patients who experienced a grades 2–4 acute diarrhea were 49% (52/106) and 44% (48/109) in the placebo and LAO groups, respectively (P = .21) (Table 2). There was also no statistically significant treatment difference within each stratum: sex, chemotherapy administration, and radiation dose (Table 3). The lack of treatment effect persisted when adjusting for prognostic factors, such as disease site, sex, age, Zubrod status score, and chemotherapy administration (odds ratio = 0.69; 95% confidence interval = 0.39 to 1.22).

Table 1.

Pretreatment characteristics for all eligible patients

Characteristic	Placebo (n = 108)		Long-acting octreotide acetate (n = 111)
Age, y
Median	61		61
Range	37–85		27–83
	No.	%	No.	%
Sex
Male	69	64	69	62
Female	39	36	42	38
Zubrod performance status
Not reported	6	6	6	5
0	89	82	83	75
1	13	12	17	15
2	0	0	5	5
Chemotherapy (planned)
Bolus	19	18	20	18
Continuous	89	82	91	82
Radiation therapy dose (planned), Gy
<50	20	19	18	16
≥50	88	81	93	84
Disease site
Anal	19	18	20	18
Rectal	89	82	91	82
Diarrhea at study entry (Common Terminology Criteria for Adverse Events version 3.0)
Not reported	1	1	0	0
None	89	82	91	82
Grade 1	15	14	17	15
Grade 2	3	3	3	3

Table 2.

Acute diarrhea adverse events based on Common Terminology Criteria for Adverse Events version 3.0

	Placebo (n = 106)*		Long-acting octreotide acetate (n = 109)*
Adverse event attribution category	No.	%	No.	%	P†
	Grades 2–4 events		Grades 2–4 events
Treatment attribution
Definite/probable/possible	39	37	41	38	.45
Unlikely/unrelated‡	12	11	7	6	.10
Attribution not reported	1	<1	0	0	—§
Total	52	49	48	44	.21
	Grades 3–4 events		Grades 3–4 events
Treatment attribution
Definite/probable/possible	20	19	20	18	.46
Unlikely/unrelated‡	9	8	5	2	.12
Total	29	27	25	23	.23

^*Two patients in each group did not have follow-up adverse event information.

^†Based on two-sample binomial test (one-sided).

^‡Reasons included bowel obstruction, colostomy, or noncompliance to treatment.

^§Two-sample binomial test (one-sided) was not applicable for this category because of small number of events.

Table 3.

Acute diarrhea adverse events by stratification factors^*

	Placebo			Long-acting octreotide acetate
Stratification factor	n†	Grades 2–4 events	%	n†	Grades 2–4 events	%	P‡
No.	106	52	49	109	48	44	.21
Sex
Males	68	33	49	67	24	36	.07
Females	38	19	50	42	24	57	.26
Chemotherapy (planned)
Bolus	19	8	42	19	7	37	.38
Continuous	87	44	51	90	41	46	.25
Cancer type
Anal	19	9	47	20	10	50	.43
Rectal	87	43	50	89	38	43	.18
Radiation therapy dose (planned), Gy
<50	20	9	45	18	8	44	.49
≥50	86	43	50	91	40	44	.21

^*Common Terminology Criteria for Adverse Events Version 3.0.

^†Two patients in each group did not have follow-up adverse event information.

^‡Based on two-sample binomial test (one-sided).

In terms of overall adverse events in the placebo group, one patient experienced treatment-related severe (grade 4) dehydration; four patients experienced treatment-related severe hematologic adverse events; and one patient died because of multiorgan failure not attributed to protocol treatment. In the LAO group, one patient experienced treatment-related severe infection; three patients experienced treatment-related severe hematologic adverse events; and two patients experienced severe neurological events not attributed to protocol treatment. No patient in either group experienced severe gallstone formation.

The secondary endpoints included the effect of LAO on treatment completion. There was no statistically significant difference between the treatment groups in study drug compliance (P = .82). A total of 170 (78%) patients completed the study drug regimen according to the protocol and received two 30-mg injections of the study drug; seven (3%) patients experienced an adverse reaction to the test dose and did not begin the study drug treatment; and 29 (13%) patients experienced other adverse events or complications, including diarrhea, dehydration, and infection, resulting in discontinuation of the treatment (Table 4).

Table 4.

Study drug, chemotherapy, and radiation therapy compliance

Treatment characteristic	Placebo (n = 108)		Long-acting octreotide acetate (n = 111)		P
Chemotherapy duration, median (range), d	37.0 (1–144)		38.0 (1–365)		.25*
Radiotherapy duration, median (range), d	40.0 (4–188)		38.0 (11–105)		.31*
	No.	%	No.	%
Study drug
Treatment completed per protocol	81	75	89	80	.82†
Adverse reaction to test dose	3	3	4	4
Other adverse events	17	16	12	10
Patient refusal	5	4	3	3
Not reported	2	2	3	3
Chemotherapy
Bolus	13	12	18	16	.14†
Continuous	93	86	86	78
Not reported	2	2	7	6
Chemotherapy dose modification
No	76	70	67	60	.27†
Yes	30	28	41	37
Not reported	2	2	3	3
Chemotherapy agent
5-fluorouracil	78	72	70	63	.45†
Capecitabine	13	12	11	10
Mitomycin-C	8	8	13	11
Other	3	3	9	8
Not reported	6	6	8	7
Upper border of initial radiotherapy field
Not reported (no radiotherapy)	3	3	5	4	.58†
Lowest sacroiliac joint	2	2	5	4
Mid sacroiliac joint	11	10	15	14
Top of sacroiliac joint	78	73	70	63
Other	14	12	16	15
Radiotherapy dose modification
No	68	62	67	60	.95†
Yes, adverse event	19	18	22	20
Yes, other reason	19	18	19	17
Not reported	2	2	3	3
Number of hospitalizations
None	82	76	92	83	.55†
1	16	15	10	9
2	5	5	5	4
≥3	3	3	1	1
Not reported	2	2	3	3
Use of other antidiarrheal agents
No	38	35	44	39	.67†
Yes	68	63	64	58
Not reported	2	2	3	3

^*Based on Wilcoxon rank-sum test (two-sided).

^†Based on Fisher exact test (two-sided).

LAO did not reduce the number of hospitalizations required (P = .55) or eliminate the need for additional standard antidiarrheal agents (P = .67). It also did not prevent modifications, delays, and interruptions in chemotherapy (P = .27) or radiation therapy (P = .95) (Table 4). Because of adverse events or complications during treatment, 18% of patients were hospitalized. Approximately 60% of the patients used antidiarrheal agents, primarily Imodium. The median chemotherapy duration was 37 days (range 1–365 days), with the majority of patients (68%) receiving 5-FU, and 32% of patients experienced chemotherapy modifications. The median radiation duration was 39 days (range 4–188 days), and 36% of patients experienced radiotherapy modifications.

One hundred sixty-three (75%) patients completed both QOL-RTI assessments, 151 (69%) patients completed both EPIC-Bowel assessments, 153 (70%) patients completed both FACE-Bowel assessments, and 125 (57%) patients completed both DAS assessments (Table 5). Patients who missed the baseline or the 3-month follow-up assessments were excluded from the analyses. The data are assumed to be missing completely at random (MCAR), and excluding patients should not lead to biased results as pretreatment characteristics did not show a statistically significant difference between patients with complete and missing data. Therefore, patients included in the analysis are a representative sample of all eligible patients.

Table 5.

Patient-reported assessment compliance^*

	All patients	Patients missing both assessments		Patients missing baseline assessment		Patients missing 3-month assessment		Patients completing both assessments
Assessment tool	N	n	%	n	%	n	%	n	%
QOL-RTI
Placebo	107	6	6	6	6	17	16	78	72
LAO	111	6	5	7	6	13	12	85	77
Total	218	12	5	13	6	30	14	163	75
EPIC-Bowel
Placebo	107	8	7	10	9	17	16	72	68
LAO	111	6	5	11	10	15	14	79	71
Total	218	14	6	21	10	32	15	151	69
FACE-Bowel
Placebo	107	8	7	7	7	19	18	73	68
LAO	111	5	5	9	8	17	15	80	72
Total	218	13	6	16	7	36	17	153	70
DAS
Placebo	107	10	9	15	14	20	19	62	58
LAO	111	9	8	16	14	23	21	63	57
Total	218	19	9	31	14	43	20	125	57

^*DAS = Diarrhea Assessment Scale; EPIC = Expanded Prostate Cancer Index Composite; FACE = Functional Alterations due to Changes in Elimination; LAO = long-acting octreotide acetate; QOL-RTI = Quality of Life-Radiation Therapy Instrument.

We did not observe a statistically significant difference between treatment groups in the proportion of patients who reported improved QoL or bowel function at 3 months (among evaluable patients) in any of the four assessments. On the QOL-RTI, 29% (47/163) of evaluable patients improved; on the EPIC-Bowel, 40% (60/151) of patients improved; on the FACE-Bowel, 25% (39/153) of patients improved; and on the DAS, 44% (55/125) of patients improved (Table 6).

Table 6.

Patient symptom scale change scores^*

		Placebo (n = 107)		Long-acting octreotide acetate (n = 111)
Assessment tool	SEM	n	%	n	%	P†
QOL-RTI	11.91
Not evaluable		29	28	26	23
Evaluable		78	72	85	77
Deteriorated‡		34	44	34	40	.47
Stable		25	32	23	27
Improved		19	24	28	33
EPIC-Bowel	11.99
Not evaluable		35	33	32	29
Evaluable		72	67	79	71
Deteriorated		20	28	29	37	.49
Stable		22	30	20	25
Improved		30	42	30	38
FACE-Bowel	2.06
Not evaluable		34	32	21	28
Evaluable		73	68	80	72
Deteriorated		21	29	26	33	.39
Stable		33	45	34	42
Improved		19	26	20	25
DAS	0.82
Not evaluable		45	43	48	43
Evaluable		62	57	63	57
Deteriorated		28	45	22	35	.54
Stable		9	15	11	17
Improved		25	40	30	48

^*DAS = Diarrhea Assessment Scale; EPIC = Expanded Prostate Cancer Index Composite; FACE = Functional Alterations due to Changes in Elimination; QOL-RTI = Quality of Life-Radiation Therapy Instrument; SEM = standard error of measurement.

^†Based on one-sided χ² test.

^‡Deteriorated: patient 3-month score at least 1.0 SEM worse than pretreatment score; stable: patient 3-month score within 1.0 SEM from pretreatment score; improved: patient 3-month score at least 1.0 SEM better than pretreatment score.

The prophylactic use of long-acting octreotide acetate did not show a statistically significant reduction in the incidence of mild, moderate, or severe diarrhea; hospitalizations; utilization of medical resources; QoL; or bowel function in patients receiving concurrent chemoradiotherapy for rectal or anal cancer. Therefore, when evaluating the relationship between the physician-assessed CTCAE diarrhea tool and the patient-assessed diarrhea tools, adjustments because of treatment were not needed. There was a statistically significant correlation between the CTCAE diarrhea grade and the EPIC (Spearman r = .26, P < .001), DAS (Spearman r = .21, P = .003), and FACE (Spearman r = .22, P = .001) total scores.

Discussion

We found that LAO (30 mg intramuscularly the week before and on day 22 of pelvic radiation) did not show a statistically significant reduction in the incidence or severity of diarrhea or change in patient-reported bowel function and QoL in patients with rectal or anal cancer undergoing curative chemoradiation. The rates of grades 2–4 diarrhea and reported loss of bowel function and QoL were nearly identical on both (placebo vs LAO) treatment groups. Our group of subjects was fairly homogeneous. The majority of patients (82%) received treatment for rectal cancer with concurrent continuous infusion of 5-FU (68%). Moreover, the compliance with LAO, placebo, and chemoradiation administration was quite high. The rates of diarrhea were also consistent with previous reported results (6,12,13).

There is lack of prospective data concerning patient-reported QoL and anorectal function following chemoradiation for both rectal and anal cancers. Overall, our investigation did not demonstrate statistically significant changes in QoL or bowel function outcomes between treatment groups. However, the majority of patients in each treatment group were categorized as either stable or improved on these measures after treatment, suggesting differential QoL and functional outcome experience. As part of a prospective randomized trial evaluating the impact of sucralfate on bowel function in patients receiving pelvic radiotherapy, Haddock et al. (42) found that all measures of bowel function worsened (deleterious impairment) during radiotherapy, as compared with pretreatment, using a nine-item patient-reported instrument. Although some measures (eg, frequency of bowel movements, cramping, and bleeding) returned to near-pretreatment baseline by 1-year posttreatment, symptoms of bowel urgency and fecal incontinence were persistent at 1 year (42). A previous study using LAO for the prevention of chemotherapy-induced diarrhea showed no differences between treatment groups in patient-reported QoL or treatment satisfaction (14). Other interventional studies have not found improvement in QoL outcomes in short-term or long-term assessments (38,43,44). These conflicting findings suggest the need for further assessments and interventions for ameliorating both acute- and long-term QoL and bowel function changes associated with cancer treatment for lower gastrointestinal malignancies.

Perhaps of greater importance is that this study supports an adequate reliability and validity of the four patient-reported assessments to the CTCAE diarrhea tools. This finding suggests that it may be appropriate to adopt these measures in future symptom management studies for rectal or anal cancer. Additional analyses with a larger sample are needed to further validate the relationship of patient-reported measures with clinician-assessed tools to improve measurement and evaluation of patient outcomes in symptom management trials.

Our overall trial results are consistent with another cooperative group's (North Central Cancer Treatment Group [NCCTG]) phase III trial that evaluated use of LAO depot in the prevention of radiation-induced diarrhea (44). In this study by Martenson et al., patients with cancer confined to the pelvis received radiotherapy to a minimum of 45 Gy, with or without 5-FU, were chosen. All patients were randomly assigned to receive 100 mg octreotide acetate, or an equal amount of placebo, via subcutaneous injection, on the first day of radiation treatment. Patients were assessed for toxic effects on the second day, and if they had no signs of toxicity, they received a dose of 20 mg LAO or placebo injection administered intramuscularly. A second intramuscular injection of 20 mg LAO or placebo was administered on day 29. Long-acting octreotide did not reduce the severity or incidence of diarrhea during pelvic radiotherapy. Grades 2–3 diarrhea was 52% and 43% in the LAO and placebo groups, respectively (P = .64). Some measures of bowel functions appeared worse in the octreotide group, including nocturnal bowel movements (70% vs 45%, P = .004), clustering bowel movements (90% vs 69%; P = .004), and bleeding with bowel movements (57% vs 35%, P = .01). No other statistically significant differences in toxicity, patient-reported symptoms, or QoL were observed. One serious concern on this trial was including a heterogeneous group of subjects consisting of rectal, prostate, and gynecologic cancers and less than 50% of the patients receiving various forms of concurrent systemic therapy. However, despite the differences with our study, their results remain consistent with ours. Therefore, we can infer from both trials that LAO is not effective in the prevention of radiation- or chemoradiation-induced diarrhea and should not be used outside of a clinical trial.

Our study may have one limitation because it did not account for the surgical status (preoperative vs postoperative) and the bowel-sparing techniques that would reduce incidence of diarrhea by moving normal bowel from the radiation field. However, Martenson et al. (44) included more postoperative rectal cancer patients in the placebo group than in the treatment group and found no differences in acute toxicity or efficacy of octreotide.

Although the use of LAO for the prevention of chemoradiation-induced diarrhea appears to be ineffective, the role of octreotide in the treatment of chemotherapy or radiation-induced enteritis remains promising (28,30,31,45). Yavuz et al. (45) assessed the effectiveness of SAO in patients who developed diarrhea from radiation in a small single institution randomized effort (45). Sixty-one patients who experienced grade 2 or 3 diarrhea during pelvic radiation were randomly assigned to receive either a dose of 0.1 mg SAO, administered subcutaneously three times a day, or an oral dose of a combination of 2.5 mg diphenoxylate and atropine. In this trial, the use of daily SAO was associated with improved control of acute radiation-induced diarrhea. Twenty of the 33 patients treated with daily octreotide had resolution of diarrhea within 3 days of starting the medication, in contrast to only four of the 28 patients treated with diphenoxylate and atropine (P = .002). Interpreting the totality of the results from our study and these two trials (44,45) is challenging because there are no studies comparing LAO and SAO for prevention of chemoradiation-induced diarrhea.

The RTOG, as a core mission of its Symptom Management Committee, has previously evaluated other agents that appeared encouraging in phase II trials for the treatment of diarrhea associated with pelvic chemoradiation. The double-blinded phase III investigation, RTOG 9809, investigated the use of pentosanpolysulfate in the treatment of gastrointestinal tract sequelae (proctitis, diarrhea, and melena) of pelvic radiotherapy (38). No gastrointestinal toxicity benefit was observed for patients receiving pentosanpolysulfate in this phase III evaluation. Currently, the RTOG is evaluating the utility of intensity-modulated radiation therapy in reducing acute gastrointestinal toxicity associated with intensified preoperative therapy in rectal cancer (46).

Funding

The trial and the publication of the results are supported in part by grants RTOG U10 CA21661, CCOP U10 CA37422, and Stat U10 CA32115 to the Radiation Therapy Oncology Group (RTOG) from the National Cancer Institute and Novartis.