RESOURCE TYPE: Expert opinion

INCLUSION CRITERIA: Ipilimumab, colitis, perforation, metastatic melanoma

The authors developed an algorithm for caring for patients who develop diarrhea with ipilimumab treatment. Three case studies were outlined. In one study, a patient developed autoimmune colitis after an infectious diarrhea workup was negative, was treated with high-dose glucocorticoids, and improved. Another patient presented with bloody diarrhea, leukocytosis, negative stool cultures, and negative C-difficile. A colonoscopy showed severe active colitis with ulcerations that were inflammatory based on biopsy and pathology. The patient was treated with infliximab and improved. The third patient presented with nausea/vomiting, and diarrhea for one week, had negative stool and C-difficile cultures, no leukocytosis, and a normal erythrocyte sedimentation rate. Autoimmune colitis was noted on a sigmoidoscopy. The patient improved with glucocorticoids.

Case study of three patients, each with a different presentation. Two of the patients were octogenarians, and both developed more severe symptoms than the third patient, who was 51 years. Patient comorbidities were not identified.

Nurses need to carefully assess gastrointestinal symptoms in patients receiving ipilimumab to minimize complications.

To evaluate the effectiveness of a combination of aprepitant, palonosetron, and dexamethasone during the acute and delayed phase of chemotherapy-induced nausea and vomiting (CINV) in patients with lung cancer receiving carboplatin-based, moderately emetogenic chemotherapy or cisplatin-based, highly emetogenic chemotherapy

• Day 1: Aprepitant 125 mg PO, palonosetron 0.75 mg IV, and dexamethasone 3.3 mg IV
• Days 2 and 3: Aprepitant 80 mg PO

• Day 1: Aprepitant 125 mg PO, palonosetron 0.75 mg IV, and dexamethasone 9.9 mg IV
• Days 2 and 3: Aprepitant 80 mg PO and dexamethasone 3.3 mg IV

• N = 133 courses
• MEAN AGE = 65.8 years
• MALES: 77%, FEMALES: 23%
• KEY DISEASE CHARACTERISTICS: Lung cancer
• OTHER KEY SAMPLE CHARACTERISTICS: Prior experiences of CINV and motion sickness and a history of alcohol consumption

• SITE: Single site    
• SETTING TYPE: Not specified    
• LOCATION: Department of Respiratory Medicine at Sasebo City General Hospital in Japan

• PHASE OF CARE: Active antitumor treatment
• APPLICATIONS: Palliative care 

This study was a prospective trial.

• Patients completed a self-reported questionnaire on the presence of emesis (yes or no), level of nausea (four choices), and status of salvage treatment. This questionnaire was completed on days 1–5.

Patients who received carboplatin or cisplatin-based chemotherapy had CINV, even with prophylaxis including aprepitant, palonosetron, and dexamethasone. There was no difference in CINV rates between patients who received carboplatin and those who received cisplatin-based chemotherapy. This study did not compare this regimen to any other prophylaxis, so it is difficult to draw a conclusion regarding the usefulness of this regimen for CINV prophylaxis for moderately or highly emetogenic chemotherapy.

• Risk of bias (no control group)
• Findings not generalizable
• Other limitations/explanation: This study only included patients with lung cancer who received carboplatin- or cisplatin-based chemotherapy. The prospective study included only one facility. The findings of this study may not be generalizable.

To determine the best antiemetic drug combinations for patients receiving moderately emetogenic chemotherapy (MEC)

Chemotherapy-naïve patients with a mix of malignancies receiving MEC were randomized to two treatment groups. Group A received palonosetron plus one day of dexamethasone and group B received fosaprepitant, granisetron, and dexamethasone on day 1. The primary endpoint was complete response (CR, no emesis and no rescue drugs). The secondary endpoints were complete control (CC, no vomiting, no use of rescue drugs, and no more than mild nausea), total control (TC, no nausea), and therapy chosen by patients. CR, CC, and TC were measured in the acute, delayed, and overall phases for five days. Data on chemotherapy-induced nausea and vomiting (CINV) were collected on days 2 and 5 following chemotherapy.

• N = 35
• AGE RANGE = 60 ± 14 years
• MALES: 37%, FEMALES: 63%
• KEY DISEASE CHARACTERISTICS: Solid malignant tumors (majority breast and colon)
• OTHER KEY SAMPLE CHARACTERISTICS: Eastern Cooperative Oncology Group performance statuses of 0, 1, and 2

• SITE: Single site
• SETTING TYPE: Outpatient
• LOCATION: Japan

• PHASE OF CARE: Active antitumor treatment

Prospective, single-blinded, randomized crossover study

• Japanese version of the ​Multinational Association of Supportive Care in Cancer (MASCC) Antiemesis Tool (MAT)
• Patients recorded the incidence and severity of CINV on days 2 and 5. The severity of nausea was recorded using a Numeric Rating Scale (NRS).

There were no significant differences in the efficacy of the two protocols, and no significant difference in CC or TC during the acute, delayed, or overall period was found. Nausea scores were not reported although their collection was reported. These results suggest that palonosetron and dexamethasone have the same efficacy as fosaprepitant, granisetron, and dexamethasone.

The combination of palonosetron and dexamethasone was as effective as triple-drug antiemetic regimens for patients receiving MEC.

• Small sample (< 100)
• Baseline sample/group differences of import
• Key sample group differences that could influence results
• Other limitations/explanation: CINV data were only collected on days 2 and 5.

The data from this study were not strong enough to affect nursing care; however, because of the similar efficacy of the study drugs, additional studies should be investigated.

To evaluate the antiemetic potential of palonosetron, aprepitant, and dexamethasone in patients with urothelial cancer receiving gemcitabine and cisplatin chemotherapy

Patients received one of two antiemetic regimens, ondansetron or granisetron plus dexamethasone, or palonosetron, aprepitant, and dexamethasone, and over one cycle of chemotherapy were evaluated for chemotherapy-induced nausea and vomiting (CINV) events, rescue medications needed, and food intake.

• N = 122   
• AGE RANGE = 36–82 years
• MEDIAN AGE = 68–70 years
• MALES: 86%, FEMALES: 13.9%
• CURRENT TREATMENT: Chemotherapy
• KEY DISEASE CHARACTERISTICS: Urothelial cancer
• OTHER KEY SAMPLE CHARACTERISTICS: Patients receiving gemcitabine and cisplatin chemotherapy

• SITE: Multi-site   
• SETTING TYPE: Not specified    
• LOCATION: Japan

• PHASE OF CARE: Active antitumor treatment

• Nonrandomized, nonblinded, two-group comparison

• CINV events and anorexia were identified through review of the medical record and rated based on criteria from the Common Terminology Criteria for Adverse Events (CTCAE), version 4.0.
• The rescue medications used were identified through the review of the medical record.

Patients in the palonosetron, aprepitant, and dexamethasone group had significantly fewer episodes of CINV, anorexia, and rescue medication used compared to the ondansetron or granisetron plus dexamethasone group during the first cycle of chemotherapy (p = 0.012) and overall during chemotherapy (p = 0.0019).

The use of palonosetron plus aprepitant and dexamethasone results in significantly fewer episodes of CINV, anorexia, and rescue medications used in people with urothelial cancer treated with gemcitabine and cisplatin as compared to another commonly used antiemetic regimen, ondansetron or granisetron plus dexamethasone.

• Small sample (< 100)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Key sample group differences that could influence results
• The two groups were different sizes: ondansetron or granisetron plus dexamethasone (n = 75), and palonosetron, aprepitant, and dexamethasone (n = 47), so the intervention effect may be influenced.

Palonosetron, aprepitant, and dexamethasone may offer more relief from CINV in people being treated with gemcitabine and cisplatin chemotherapy.

To examine the efficacy of family focused grief therapy on psychosocial functioning in families of patients who are terminally ill

Patients and relatives in several palliative care centers and hospices were recruited for the study. Families were randomly assigned to receive the focused grief therapy or usual care stratified based on recruitment site. Usual care consisted of standard palliative care provided by homecare programs, which involved counseling when deemed clinically appropriate. Focused grief therapy was provided by qualified family therapists who had received standardized training in the intervention. Clinical supervision of the therapists was provided throughout the trial, and fidelity to the intervention was independently evaluated from review of audiotaped sessions. Family focused grief therapy typically included four to eight sessions of 90 minutes provided across 9–18 months. It included exploration of family cohesion, communication, and handling of conflict. Families were assigned to functional classes: dysfunctional (sullen or hostile) or intermediate. Data at baseline and follow-up at 6 and 13 months postbereavement were obtained from relatives by a research assistant.

• The sample was comprised of 81 families in the intervention group (n = 232) and 28 families in the control group (n = 130).
• Mean participant age was 42 ±16 years: mean patient age was 57 years, mean spouse age was 56 years, and mean age of children was 29 years.
• The sample (all participants) was 54% female and 46% male.
• The most common cancer types were breast, lung, brain, and a mix of types.
• All patients were in palliative care or hospice programs and had a prognosis of six months.
• Median length of time from illness diagnosis to death was 25 months, and median survival from study entry was 96 days.
• Of the families, 51% were classified as intermediate functioning, 26% were designated as sullen, and 23% were categorized as hostile.
• Ninety-five percent of families had two or more children.

• Multisite
• Palliative or hospice care setting
• Eight different countries

A randomized controlled trial design was used.

• Family Environment Scale
• Family Relationships Index
• Family Assessment Device
• Brief Symptom Inventory
• Beck Depression Inventory (cognitive items)
• Social Adjustment Scale
• Bereavement Phenomenology Questionnaire

Among all participants, those receiving the study intervention had significantly greater change in mean score of the Brief Symptom Inventory (BSI) (0.12, p = 0.02). BSI showed a nonsignificant improvement at 13 months in the intervention group than in the control group. Grief phenomena diminished similarly in both groups. There was no significant difference in social adjustment in both groups. There were no other significant differences between groups. Among family members who were most distressed, differences were greater, with 0.83 improvement in the BSI (p < 0.01), Beck Depression Inventory (p < 0.01), and the Bereavement Phenomenology Questionnaire (p = 0.05). In both the intervention and control groups, the general patterns of change in outcomes measures showed overall decline in symptoms over time. Families with intermediate functioning who received the intervention had a larger reduction in conflict level at six months than intermediate families in the control group (p = 0.03). Hostile families that received the intervention deteriorated more than hostile control families over the 13 months of bereavement (p = 0.001). Intermediate families received an average of 7 sessions, sullen families received an average of 6.4 sessions, and hostile families received an average of 9.4 sessions.

Family focused grief therapy appeared to have some benefit for the most distressed family members in terms of reduction of symptoms; however, these changes were not accompanied by improvement in family functioning. This intervention may protect against pathologic grief in highly distressed individuals. Among hostile families, the intervention was counterproductive

• The study had a small sample (< 100).
• The study had many missing assessments (61 members) and a high refusal rate.
• Within the subgroup analysis of the most distressed individuals, only 20 families were represented. It is not clear how many were in the control versus the intervention groups—differences in scores reported may be greatly affected by differences in group size.
• The total number of individuals in this analysis was not provided.
• It is unclear if some of the measures were sufficiently sensitive to change in order to demonstrate response to interventions.
• Results among hostile families point to the potential for harm associated with this type of intervention.
• The study was conducted across nations that are very different (e.g., Australia, United Kingdom, Africa). It is not clear how cultural context could have affected findings, and distribution of cases did not enable relevant subgroup analysis.
• In the intervention group, 24% of families withdrew from the study, suggesting some degree of potential sample bias in the results. Functioning level of withdrawals was not reported.

Family focused interventions to mitigate grief issues may be helpful in some families for highly distressed individuals; however, among the most dysfunctional families, this type of intervention might be counterproductive. Additional research is needed to determine the appropriate role of this therapy approach in palliative care.

The intervention was cognitive-existential therapy (CEGT) provided in nine Australian hospitals. Existential themes of anxiety about death and uncertainty were incorporated into six goals of therapy: promoting a supportive environment, facilitating grief work over multiple losses, altering maladaptive cognitive patterns, enhancing problem-solving and coping skills, fostering a sense of mastery, and sorting out priorities for the future. The CEGT group had 20 weekly sessions, 90 minutes each, over six months. The control group had three 50-minute relaxation classes using progressive muscle relaxation with guided imagery. Measurements were taken at baseline, 6 months, and 12 months after the intervention. The intervention was offered by 15 therapists recruited from psychiatry, psychology, social work, occupational therapy, and oncology nursing staff. All therapists received specialized training and supervision through a series of workshops using 68-page manual.

• The sample was comprised of 303 women with stage I or II breast cancer stratified by nodal status, hormone receptor status, and tumor size.
• The intervention group (n = 154) received CEGT plus three relaxation classes.
• The control group (n = 149) received three relaxation classes.

Nine Australian hospitals

A randomized controlled trial/longitudinal study design was used.

• Monash Interview for Liaison Psychiatry (structured psychiatric interview validated with DSM-IIIR)
• Affect Balance Scale
• Hospital Anxiety and Depression Scale (HADS)
• Mental Adjustment to Cancer Scale (MAC)
• Family Assessment Device
• Satisfaction with therapy and other treatments

• Baseline screening showed that one-third of the sample suffered from a form of depressive disorder.
• The CEGT intervention group had reduced anxiety (p = 0.05, two-sided) compared to controls.
• Overall effect size for group intervention was small (d = 0.25).

• The CEGT model is recommended for use in patients with early breast cancer, and the supportive-expressive treatment model is recommended for patients with advanced breast cancer.
• The psychologist intervention group had a moderate mean effect size (d = 0.52); training and experience of the therapist make the intervention more effective.

• Both groups received three relaxation classes (minor design flaw).
• The intervention required specialized training needs for therapists.

To determine whether naproxen could prevent or decrease the incidence and/or severity of pegfilgrastim-induced bone pain

A baseline assessment questionnaire was administered prior to treatment asking the presence, location, and duration of bone or joint pain, as well as how the pain was treated. A second set of questions asking whether there was development of new bone or joint pain, location, onset, duration, and severity of the effect of the assigned medication was given to patients at the time of pegfilgrastim administration. Any additional analgesia taken by the patient was also recorded. Questionnaires were completed at home and mailed. Patients were telephoned at home as a reminder to mail in their questionnaires. Enrolled patients received 500 mg of naproxen PO or placebo instructed to take prior to pegfilgrastim and continue BID. Patients were then asked to record pain severity on a 0–10 scale and duration in a daily diary. Serious adverse events (SAEs) were reported to local investigators at the University of Rochester Cancer Center Community Clinical Oncology Program research base. Other symptoms were recorded but not reportable (e.g., sleep disturbance, pain other than bone pain, fatigue).

• The study reported on 510 patients.
• The sample was 86% female and 14% male.
• Patients had a diagnosis of nonmyeloid cancer, including breast (67%), lung (10%), gynecologic (6%), hematologic (7%), and other cancers (7%).
• Patients were 18 years of age or older and were able to understand English. 
• Patients were excluded from the study if they were pregnant or nursing or had active gastrointestinal (GI) bleeding, history of GI bleeding, history of gastric or duodenal ulcers, heart surgery in the past six months, allergy to nonsteroidal anti-inflammatory drugs (NSAIDs) or aspirin, and a creatinine level more than 1.5 times the upper limit of normal. They could not be taking NSAIDs or therapeutic dose of aspirin or warfarin.

• Mutlisite (multicenter trial)
• Setting not specified

• Patients were undergoing active treatment.
• The study has clinical applicability for elderly care; late effects and survivorship; and end-of-life and palliative care.

The study was a phase III, double-blind, placebo-controlled, randomized trial.

• Pain questionnaire
• Standard symptom inventory
• National Cancer Institute’s Common Terminology Criteria for Adverse Events
• Daily diary rating pain (0–10 scale), severity, and duration

African Americans experienced more bone pain than white patients. The naproxen group showed improvement in pain reduction versus placebo. Area under the curve for pain was 7.71 for the placebo group and 6.04 for the naproxen group (p = 0.037). Naproxen decreased maximum pain from 3.40 to 2.50 (p = 0.005), incidence from 71.3% to 61.1% (p = 0.020), duration from 2.40 to 1.92 (p = 0.009), and severe pain from 27% to 19.2% (p = 0.048). Bone pain reached maximum at day 3, where naproxen patients experienced the most benefit. Patients receiving naproxen took less prescription and nonprescription pain medication. Six SAEs occurred in the naproxen group, and six SEAs occurred in the placebo group, with all SAEs unlikely or determined to be unrelated to the intervention.

There is a high incidence of pain in patients receiving pegfilgrastim. Naproxen seems to decrease incidence, duration, and severity.

The pain questionnaire was not presented in the article, and further studies may not reproduce similar results. Patients were called at home to complete the survey; prescription and nonprescription medications that were taken in addition to the naproxen and timing of the survey may have impacted results. It is also possible that the individual responding to the call may not have been the study participant. The study does not address whether individuals with other existing pain conditions were in the study or excluded or whether there were other pain conditions present at the start of the study that may have impacted results.

This study indicates there a is a difference in pain experience with different races, which may impact practice in terms of treating pain and the patient population being treated. Additional research and education are needed to investigate this difference. These results may be helpful in treating some patients’ pain; however, more work is needed to find treatments for those who cannot seek this treatment or who derive limited or no benefit from it. More than 60% of patients in this study still experienced pain, with approximately 20% experiencing severe pain. It seems this study has a good start to recognizing the existence of pegfilgrastim-induced pain and has tested a somewhat helpful intervention. However, there is still a need for more studies with different approaches to pain control and factors to predict incidence, severity, and ability to prevent pegfilgrastim-induced bone pain. The benefit of using naproxen is that it is low cost and has very few SAEs, which will need to be considered in future studies.

To evaluate the benefit of hyaluronic acid in management of radiodermatitis

Women who developed a grade 1 dermatitis with radiotherapy were randomized to use either hyaluronic acid cream or a simple urea-based emollient once daily. Patients were advised to shower one-to-two times daily. Clinical evaluation of the skin was done weekly by the radiation oncologist. Pain and quality of life were assessed at the end of treatment on day 30. The study endpoint was a success as defined by disappearance of erythema 30 days after its occurrence.

• The study sample (N = 200) was comprised of female patients with breast cancer.
• Median age was 53 years, with a range of 27–83 years.
• Median Gy was 28, with a range of 10–52 Gy.

The study used a randomised open-label phase III design.

• The Radiation Therapy Oncology Group and European Organisation for Research and Treatment of Cancer radiation toxicity scale was used.
• Erythema was measured using calorimetry.

Overall, 36.5% of participants withdrew prematurely because of worsening of eipthelitis, patient refusal, change in treatment, or use of another product. In the intent to treat population (all registered), there was a 27.3% treatment failure with hyaluronate and a 38.8% treatment failure with emollient. Power analysis showed that 100 per study group were needed. With drop outs, this study was underpowered.

 The study did not demonstrate any significant differences in epithelitis between those using hyaluronic acid or emollient topically. The study was insufficiently powered to provide any firm conclusions.

• The baseline sample differences were of import.
• The study had a risk of bias due to no blinding and sample characteristics.
• The hyaluronic acid group had a higher skin temperature at erythema area baseline. 
• Patient compliance with protocol use is not discussed, so it is not clear if use was consistent. 
• No subgroup analysis was done based on radiation therapy dosage or combined chemotherapy and radiation therapy.

Study findings are inconclusive because of the study limitations. Findings suggest that hyaluronic acid and urea-based emollients topically may provide similar results for prevention of low-grade radiodermatitis.  These findings should be viewed with caution because of the study design limitations and lack of sufficient sample size to meet power requirements.

To review the effects of guided imagery on patients with cancer experiencing pain

The type of report was systematic review.

Databases searched were PubMed, CINAHL, PsycINFO, and Cochrane Library.

Search keywords were guided imagery, cancer pain, and systematic review.

Studies were included if they were review articles published in English since 1985; the search was not limited to randomized controlled trials (RCTs).

Studies were excluded if they were clinical trials or systematic reviews that did not utilize guided imagery as an intervention, did not specifically investigate cancer pain, were not a clinical study but rather a summary of guided imagery, had a qualitative design, and were not conducted within the study time frame.

In order to focus on the most current research, this review targeted articles published during 2001–2008.

• A final number of five clinical trials were identified that included pain as either a primary or secondary outcome measure.
• Sample range across studies was 40–66 patients.
• The sample included three RCTs, one group pre/post-test design, and one RCT crossover design.

Five studies included pain as either a primary or secondary outcome measure. In three of those, pain intensity and pain-related distress decreased in the guided imagery intervention versus control for pain intensity and pain-related distress, average pain score decreased, and there was a decrease in body discomfort.

It is difficult to give concrete recommendations that guided imagery will work for all patients who suffer from cancer pain. However, based on the information from these reviews, guided imagery can be recommended as a potential aid in the relief of pain associated with cancer.

Inconsistencies and limitations included the small sample size, different patient populations, different scripts, and frequency of medication administration.

There is inconsistency in the methodological qualities of these trials. Further research is necessary to provide better evidence for the use of guided imagery in cancer pain.