To determine whether adding olanzapine to current standard antiemetic therapy could reduce the frequency of chemotherapy-induced nausea and vomiting (CINV) and improve patients’ quality of life (QOL) during chemotherapy

• N = 44 
• MEAN AGE = 63 years (olanzapine group), 55 years (control group), non-significant difference p = .06
• MALES: 50%, FEMALES: 50%
• KEY DISEASE CHARACTERISTICS: Cancer patients receiving highly emetogenic chemotherapy (HEC) or moderately emetogenic chemotherapy (MEC) with an Eastern Cooperative Oncology performance status of 0–2.
• OTHER KEY SAMPLE CHARACTERISTICS: Patients had not experienced NVR in the 24 hours before the start of the trial, had not scheduled to receive concurrent abdominal radiation therapy, had no history of DM, were not receiving other antipsychotic agents, and were not hypersensitive to olanzapine.

• SITE: Multi-site  
• SETTING TYPE: Inpatient  
• LOCATION: Sapporo, Hokkaido, Japan

• PHASE OF CARE: Active antitumor treatment

Randomized, double-blind, placebo-controlled trial

• Primary endpoint: Total control (no vomiting, no use of rescue medications, and maximum nausea of < 5/100 mm)
• Secondary endpoint: Functional Living Index–Emesis (FILE) questionnaire scores on days 0 and 6
• Additional: QOL (FLIE), amount of diet intake during CTx, satisfaction, complete response (no vomiting, no use of rescue medication), complete control (complete response + nausea < 25 mm/100 mm), and Visual Analog Scale

The olanzapine group achieved better total control (59% overall, 86% in the acute phase, and 64% in the delayed phase) than the control group (23% overall, 55% in the acute phase, and 23% in the delayed phase). The olanzapine group also achieved better complete protection and complete response except for acute phase complete response (p < .05). Furthermore, the olanzapine group experienced better QOL (p < .01), and the olanzapine group indicated that CINV did not affect their daily activities whereas 36% of the control group reported influence of CINV on daily life activities. There were significant differences between the VAS for nausea and satisfaction scores for additional medication. Most of the olanzapine group patients (91%) wished to receive same protocol for future chemotherapy. Dietary intake was better maintained by the olanzapine group with a significant difference on day 2 and days 4–6.

The addition of 5 mg/day of oral olanzapine to standard therapy can reduce the frequency of CINV and improve QOL for patients receiving highly or moderately emetogenic chemotherapy.

• Small sample (< 100)
• Baseline sample/group differences of import
• Risk of bias (sample characteristics)
• Other limitations/explanation: Age of the olanzapine group was higher than the control (which could have influenced less development of CINV), and other risk factors of the CINV were not taken into consideration. MEC group also received NK1, which is not a standard antiemetic. Combination of palonosetron and olanzapine need to be investigated further, with stratification of emetogenicity of chemotherapy regimen, and the use of first and second generation 5-HT3RA between the two groups (as palonosetron 0.75mg was used in the study).

The addition of olanzapine to standard antiemetics, such as 5-HT3RA, steroids, and NK1RA, could achieve better control of CINV, especially for the delayed phase, with additional benefit in terms of QOL and less change in dietary intake. However, caution needs to be exercised in interpreting the result as the study allowed NK1 for the MEC, had age difference between the olanzapine and control group, and did not take risk factors of CINV into consideration, and the palonosetron dose of 0.75mg was higher than antiemetic recommendations.

To evaluate the safety, efficacy, and pharmacokinetic profile of a 12.5 mcg/hour transdermal matrix fentanyl patch used to manage the persistent pain of patients with cancer

On study day 1, the patch was applied. It was replaced every 3 days for a 10-day period. All patients initially received the 12.5 mcg/hour  dose. At the time of patch replacement, the dose could be increased according to pain intensity and rescue medication used. Use of opioid receptor antagonist analgesics and other opioid medication was prohibited. Patients received fast-acting morphine for breakthrough pain. Serum fentanyl levels were measured on days 4, 7, and 10. Physicians assessed global effect on day 10 or at the time of study discontinuation. Patients provided a global assessment on days 1, 4, and 7.

• The sample was composed of 85 patients.
• Of all patients, 75% were age 60 or older. The age range was 40 to older than 70.
• Of all patients, 47% were female and 53% were male. 
• Diagnoses included gastrointestinal cancer (which 40% of patients had), gynecologic cancer (10.6%), and hepatobiliary cancer (5.9%). The sample contained a mix of other cancer sites. At baseline, 69.4% of patients were taking oxycodone, 29.4% were taking morphine, and 1.2% were undergoing fentanyl injections for pain control.

• Single site
• Outpatient
• Japan

Open-label prospective trial

• Five-point scale (5 = very satisfied, 1 = very dissatisfied), to measure patient's global assessment of pain
• Dichotomous scale (1 = effective, 2 = ineffective), to measure physicians' global assessment
• Eastern Cooperative Group (ECOG) performance status

In 78 patients, a total of 316 adverse events occurred. The most frequent adverse events were nausea (which 36% of patients experienced), somnolence (30.2%), vomiting (25.6%), diarrhea (19.8%), constipation (16.3%), pyrexia (12.8%), and insomnia (10.5%). Three patients experienced evere adverse events: dyspnea and respiratory failure. Authors observed 12 cases of skin irritation at the patch site. Of all patients, 96% reported that they were satisfied or very satisfied with the 12.5 mcg/hour transdermal matrix fentanyl patch. According to global assessment by physicians, the 12.5 mcg/hour transdermal matrix fentanyl patch was as effective as the other treatments in 98% of cases.

The 12.5 mcg/hour transdermal matrix fentanyl patch appears to be a reasonable means of controlling persistent cancer pain in patients switching from low-dose morphine or oxycodone.

• The study had a small sample, with fewer than 100 participants.
• The study had a risk of bias due to no control or comparison group.
• Authors provided no specific severity rating of adverse effects.
• Distribution of final fentanyl doses associated with efficacy was not provided.
• Authors did not report how adverse events were defined or recorded.
• Authors did not report the use rescue medications.
• Authors stated the results of a visual analog scale over time; however, they did not describe the scale or how it was used.

No firm conclusions can be drawn from this study.

To determine the impact of palonosetron on chemotherapy-induced nausea and vomiting (CINV) in patients with lymphoma receiving CHOP (cyclophosphamide, hydroxydaunorubicin, oncovin, prednisone or prednisolone) chemotherapy

This single-arm study evaluated palonosetron in patients with lymphoma. Patients received 0.75 mg palonosetron as an IVP 30 minutes prior to chemotherapy. No antiemetics were allowed within 48 hours prior to treatment. Patient were further restricted from taking any antiemetics unless they experienced symptoms of CINV for 168 hours after treatment. There is no mention of other antiemetics that were used. It is unclear if the data were collected beyond one cycle. Patients were chemotherapy naïve but could have received rituximab.

• N = 54/50   
• AGE = 23–75 years
• MALES: 46%, FEMALES: 54%
• CURRENT TREATMENT: Chemotherapy
• KEY DISEASE CHARACTERISTICS: Non-Hodgkin lymphoma  
• OTHER KEY SAMPLE CHARACTERISTICS: Stage of disease, alcohol use, smoking, motion sickness, morning sickness, previous radiation

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

PHASE OF CARE: Active antitumor treatment

Single-arm study conducted at five hospitals

Patients maintained an assessment diary from 0–168 hours after chemotherapy. Questionnaires also assessed drinking history, motion sickness, pregnancy history, and emesis during pregnancy. The primary end point was the occurrence rate of nausea, vomiting, and anorexia, which were measured using the Common Terminology Criteria for Adverse Events (CTCAE), version 4.0. Secondary end points evaluated these symptoms during the acute and delayed phases of CINV, as well as frequency of rescue medication use. The patients’ physicians scored the CTCAE based on the patients' diary.

Four patients became ineligible because of steroid administration dosing and new central nervous system metastasis. Overall occurrence of nausea and anorexia of any grade for the overall phase was 56% for nausea, 62% for anorexia, and 12% for vomiting. Six patients experienced vomiting. Half of these patients experienced vomiting after completion of the five days of steroids (part of the chemotherapy regimen). Patient characteristics were evaluated using univariate analysis, which did not show any association between patient parameters and nausea or vomiting in the overall phase. Anorexia, however, was significantly associated with both sex (p < 0.001) and drinking history (p < 0.001).

Females were more likely to experience anorexia, and should be considered for NK-1 inhibitors. There was an association of nausea in the acute phase leading to the development of delayed anorexia. Clinicians should also be more attentive to development of CINV in the delayed phase after 120 hours.

• Small sample (< 100)
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Risk of bias (no appropriate attentional control condition)
• Findings not generalizable
• Provider scoring diaries, with possible prejudice

Nurses need to consider delayed CINV in CHOP chemotherapy, particularly after patients have completed the steroid component of their treatment.

To assess the efficacy of triple drug antiemetic prophylaxis for patients receiving moderately emetogenic chemotherapy (MEC)

Patients receiving MEC containing carboplatin were treated with standard triple drug antiemetic therapy of palonosetron, dexamethasone, and aprepitant. Patients were assessed from the beginning of chemotherapy to day 7.

• N = 90   
• MEDIAN AGE = 69 years
• AGE RANGE = 38-82 years
• MALES: 80%, FEMALES: 20%
• KEY DISEASE CHARACTERISTICS: Not provided
• OTHER KEY SAMPLE CHARACTERISTICS: All patients were chemotherapy naïve except for patients with lung cancer receiving tyrosine kinase inhibitors.

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

PHASE OF CARE: Active antitumor treatment

Prospective, observational

• Functional Living Index-Emesis (FLI-E) questionnaire
• 100 mm visual analog scale (VAS) for quality of life and nausea severity
• Common Terminology Criteria for Adverse Events (CTCAE)

No patient vomited within the first 24 hours after chemotherapy. For the delayed and overall phases, completed response was seen in 91.9% of patients. Complete control was seen in 88.9%–97.8% across study days. The lowest rate of complete control was seen on day 3. Men tended to have a high prevalence of complete response and complete control. Hypertension of grade 3, which may have been related to the study drugs, was seen in five patients. Comparison to findings from five other studies showed high complete response rates in the present study.

Triple drug therapy for the prevention of chemotherapy-induced nausea and vomiting (CINV) was shown to be very effective for patients receiving MEC.

• Small sample (< 100)
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Unintended interventions or applicable interventions not described that would influence results
• No discussion of any use of rescue medication
• No direct comparison to more standard MEC regimens for antiemesis

Triple antiemetic therapy is recommended in several professional guidelines for highly emetogenic chemotherapy but not for MEC. This study showed better efficacy for CINV control compared to some other studies in CINV management for MEC. Triple drug antiemetic prophylaxis should be considered for patients receiving MEC.

To evaluate the efficacy of triple antiemetic therapy consisting of a 0.75 mg dose of palonosetron, a three-day course of aprepitant, and four days of dexamethasone in the control of chemotherapy-induced nausea and vomiting (CINV) among chemotherapy-naïve patients with lung cancer undergoing highly emetogenic chemotherapy (HEC) regimens

From September 2010 to June 2012, patients received triplet antiemetic therapy (0.75 mg palonosetron IV + 9.9 mg dexamethasone IV + 125 mg of aprepitant orally) on day 1, followed by 80 mg aprepitant on days 2 and 3, and 8 mg dexamethasone on days 2–4.

• N = 67 (later excluded 4, N = 63)              
• MEDIAN AGE: 64 years (range = 36–78 years)
• MALES: 65.7%, FEMALES: 34.3%
• KEY DISEASE CHARACTERISTICS: Chemotherapy-naïve patients with lung cancer scheduled to receive HEC [50 mg/m² or higher dose of cisplatin with concurrent agents (pemetrexed [38.8 %], vinorelbine [32.7 %], gemcitabine [9.0 %], irinotecan [7.5 %], etoposide [7.5 %], and docetaxel [4.5 %])
• OTHER KEY SAMPLE CHARACTERISTICS: No emesis within 24 hours of chemotherapy administration; no symptomatic or suspected brain metastasis; no concomitant radiotherapy; no complications that prohibited dexamethasone use; and no known hypersensitivity to palonosetron, aprepitant, or dexamethasone

• SITE: Multi-site  
• SETTING TYPE: Outpatient  
• LOCATION: Japan

• PHASE OF CARE: Active antitumor treatment

Single-arm, phase II prospective, multi-centered, longitudinal trial

• CR rates: 81% (overall), 96.8% (acute), and 81.0% (delayed)
• CC rates: 63.5% (overall), 92.1% (acute), and 63.5% (delayed)
• No nausea: 54% (overall), 84.1% (acute), and 54.0% (delayed)
• No significant nausea: 66.7% (overall), 93.8% (acute), and 66.7% (delayed)
• More than half of the patients suffered from appetite loss due to CINV.
• No of adverse events exceeded grade 3 of the Common Terminology Criteria for Adverse Events (CTCAE). Constipation was the most frequent adverse event.

The triplet antiemetic therapy was the most promising regimen in preventing CINV in chemotherapy-naïve patients with lung cancer treated with HEC. The dose-response relationship between palonosetron and aprepitant use needs to be further investigated. However, the experience of nausea by almost half of the patients during the overall phase highlighted the necessity of conducting further investigations to control CINV during the delayed phase.

• Small sample (< 100)
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Measurement validity/reliability questionable
• Findings not generalizable
• No consideration about CINV risk factors such as the explanation not measuring for patients’ history of alcohol consumption; experiencing nausea and vomiting during previous radiotherapy or during pregnancy for female patients; and anxiety level. As the starting of the study was on the first cycle of chemotherapy, the investigators had no chance to determine prior experience of CINV and thus anticipatory nausea and vomiting.

Although adding palonosetron to antiemetic regimens shows improvement more impressively in the acute phase among patients, caution should be taken before interpreting the result of this study as it did not use the most robust research design (of randomization, blinding, placebo control) to detect the drug's efficacy. Constipation is a commonly reported adverse effect in this study, therefore careful management should be considered when palonosetron is used.

• FINAL NUMBER STUDIES INCLUDED = 18; 3 were conference abstracts 
• TOTAL PATIENTS INCLUDED IN REVIEW = 22,882
• SAMPLE RANGE ACROSS STUDIES: 20–32,072
• KEY SAMPLE CHARACTERISTICS: Multiple tumor types

PHASE OF CARE: Active antitumor treatment

Findings suggest that the use of long-acting CSFs was associated with a lower risk of FN, FN hospitalizations, and other related adverse patient outcomes.

• Tumor types varied. Those with hematologic cancers were included, but there was no subgroup analysis to differentiate from those with lower-risk solid tumor types.
• Variability of chemotherapy regimens involved
• Variation in the timing and duration of short-acting CSF interventions and dose differences that were not captured in the data
• Known potential inaccuracies of medical claims data from which most cases were obtained
• There is no single diagnosis code for FN, and the algorithms used to define FN can vary.
• Low GRADES quality rating of studies included.
• Some data was provided per patient, and some was provided per cycle.
• In some studies, patients had received both filgrastim and pegfilgrastim.

Findings suggest that use of long-acting CSF may provide better prophylaxis for FN and FN-related events, but data and limitations of this review were insufficient to draw firm conclusions. Because of the variety of new biosimilar alternatives, additional work to determine comparative effectiveness and cost-benefit analysis of various formulations is increasingly important. This study does raise the question of what denominator is best used in outcomes measurement, whether outcomes should be viewed per patient, per cycle, or per patient-cycle.

To determine the efficacy of alleviating fatigue using the psychostimulant methylphenidate hydrochloride (MPH) in patients with advanced cancer

MPH 5 mg was taken orally twice daily versus placebo for three cycles of a pair of three-day periods. Patients completed a daily diary of symptom scales and side effects.

• N = 24  
• MEDIAN AGE = 71 years
• MALES: 54%, FEMALES: 46%
• KEY DISEASE CHARACTERISTICS: Gastrointestinal, prostate, lung, breast, and genitourinary cance
• OTHER KEY SAMPLE CHARACTERISTICS: ​Australia-modified Karnofsky Performance Scale (AKPS) ≥ 40; screening fatigue score ≥ 4/10 (National Comprehensive Cancer Network); stable treatment regimen for at least 48 hours; no plan for treatment likely to influence fatigue; no change in thyroxine, antidepressant therapy, or other drugs with sympathomimetic potential for three weeks prior to recruitment

• SITE: Multi-site  
• SETTING TYPE: Not specified    
• LOCATION: Queensland and New South Wales, Australia

• PHASE OF CARE: End-of-life care
• APPLICATIONS: Palliative care

Population was estimated using the aggregated N of one multicycle, double-blinded, controlled, crossover study.

• Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F)
• Wu Cancer Fatigue Scale (WCFS)
• Edinburgh Depression Scale (EDS)
• AKPS performance scale

Eight patients had individual improvements in fatigue with MPH compared to a placebo on the FACIT-F and WCFS scores, but the mean population estimate showed no important difference. Seven patients showed an improvement in EDS scores, but the mean population estimate showed no important difference. There was no change in AKPS scores. There were six adverse events with three events possibly related to MPH.

Although there may be some individual improvement in fatigue with MPH use, the results of this small sample size were difficult to generalize.

• Small sample (< 30)
• Selective outcomes reporting
• Measurement validity/reliability questionable
• Subject withdrawals ≥ 10%  
• Other limitations/explanation: Patients self-reported; N of one design; intention to treat analysis used a combination of weighted group estimates and prior values for missing data; of 43 patients originally recruited, only 55.8% completed three cycles of treatment; the mean population response estimate also included patients who did not complete any study cycle, so it is unclear if results are over or underestimated; study cycle times were short and may not have allowed for full drug effects

Continued studies on the effects of psychostimulants on cancer-related fatigue are needed. This study adds to the growing body of evidence that methylphenidate is not generally helpful in reducing fatigue. The strength of this study's results are limited by design issues.

To test whether a general practice consultation directed by a carer needs checklist would improve meeting the needs of carers

General practitioners of intervention patients were visited, introduced to intervention resources, and invited to participate. Tools provided were a needs assessment tool and a general practice toolkit of paper-based and electronic forms providing evidence-based information, resources, and services that might help address problems. Consultations were provided at baseline and at three months, based on needs assessment results. Patients were randomized, but those whose general practitioner had a previous participant were nonrandomly allocated to the same study group to prevent contamination. Data collection included survey at one, three, and six months by telephone interview. Control group general practitioners were not contacted. General practitioners were educated on the process through academic detailing.

• N = 211 carers
• MEAN AGE = 57.5 years
• MALES: 33.5%, FEMALES: 66.5%
• KEY DISEASE CHARACTERISTICS: Varied prognoses in terms of life expectancy 
• OTHER KEY SAMPLE CHARACTERISTICS: Most carers were spouse or partner and lived with the patient. 32% were employed full-time, and 18.5% were employed part-time. 

• SITE: Multi-site  
• SETTING TYPE: Outpatient  
• LOCATION: Australia

• PHASE OF CARE: Late effects and survivorship
• APPLICATIONS: Palliative care 

• Single-blind RCT

• Supportive care needs survey for partners and carers to measure unmet needs
• Hospital Anxiety and Depression Scale
• SF-12

Twenty-nine percent of intervention group patients and 15% of control group patients dropped out of the study—most said this was because they were too busy to continue after their initial interview. Three of 158 general practitioners refused to conduct the consultations. No between-group differences were seen in change from scores at baseline to scores at any time point. Subgroup analysis showed no change in depression scores. For those who were clinically depressed at baseline, control patients demonstrated significantly worse anxiety at six months, while intervention group scores were essentially stable. Intervention group carers with baseline anxiety or depression reported deterioration in physical scores on the SF-12 (p = .053). In analysis adjusted for baseline anxiety, those in the intervention group had significant worsening of physical function scores (p = .037) and increased psychological and emotional needs from baseline to three months (p = .033). Those caring for less ill individuals had improvement in mental health at three months in SF-12 scores.

The study did not demonstrate improvement in intensity or number of unmet needs or carer outcomes overall.

• Risk of bias (no random assignment) 
• Risk of bias (no appropriate attentional control condition)  
• Unintended interventions or applicable interventions not described that would influence results
• Questionable protocol fidelity 
• Subject withdrawals 10% or greater 
• Other limitations/explanation: Although general practitioner participation was stated, actual completion of expected consultations at the expected timeframes is not discussed or reported—no approach to ensure treatment fidelity in such consultations; high attrition rate; no information about the patient’s type of cancer, severity, or symptoms is provided—these are likely to have an impact on the degree of caregiver burden and associated outcomes; the nature of the general practice consultation is not clearly described, and actions taken to meet assessed needs is not discussed.

Findings did not support the idea of reducing carer unmet needs via an assessment-driven general practice consultation over time. Undergoing assessment of needs possibly increases the degree to which carers identify needs that are not being met. This study focused on provision of tools to general practitioners but does not describe what actions were taken on the basis of the assessment done, so it provides little support for any specific approach to address caregiver needs.

To determine if prophylactically administered 600 mg oral gabapentin reduces postoperative nausea or emesis and decreases postcraniotomy pain

Patients to undergo elective craniotomy were randomized to receive either placebo (a vitamin b-complex capsule) or 600 mg oral gabapentin two hours before induction of anesthesia. All received standard antiepileptic prophylaxis with 100 mg phenytoin every eight hours and 4 mg IV dexamethasone every eight hours for 48 hours prior to surgery. Patients were given 1 gm IV paracetamol every six hours for postoperative pain. Rescue analgesia was provided with 1 mcg/kg IV fentanyl for a pain score of 3. Rescue antiemetic was provided with 4 mg IV ondansetron if the patient had any emetic episode or 10 minutes or longer of nausea. All received the same anesthesia. Symptoms were assessed hourly for the first six hours postoperatively and then every two hours for the next 18 hours.

• N = 73  
• MEAN AGE = 44 years
• MALES: 68.5%, FEMALES: 31.5%
• KEY DISEASE CHARACTERISTICS: All were undergoing craniotomy
• OTHER KEY SAMPLE CHARACTERISTICS: Duration of anesthesia averaged slightly more than 400 minutes and was not different between study groups.

• SITE: Single site   
• SETTING TYPE: Inpatient   
• LOCATION: India

• PHASE OF CARE: Active antitumor treatment

• Placebo-controlled RCT

• Pain verbal rating scale of 0 (no pain) to 3 (moderate to severe, requiring rescue)

Incidence of nausea was 35.1% in the placebo group compared to 11.1% in the gabapentin group (p = .02). No significant difference was seen between groups in incidence of emesis, but there was a trend to lower incidence of emesis with gabapentin. No difference was seen between groups in postoperative pain scores, the number of patients who required rescue analgesia, or postoperative fentanyl consumption.

Prophylactic oral gabapentin prior to surgery reduced postoperative nausea and vomiting in patients undergoing craniotomy. Perioperative gabapentin had no effect on postoperative pain.

• Small sample (less than 100)
• Risk of bias (no blinding)
• Measurement validity/reliability questionable
• The method of pain measurement is questionable, as a 3-point scale.

Preoperative gabapentin may reduce postoperative symptoms of nausea and vomiting in patients undergoing craniotomy who are receiving dexamethasone perioperatively. The optimum dosage of gabapentin has not been determined.

To compare the efficacy of amitryptylline, gabapentin, and pregabalin in patients with cancer experiencing neuropathic pain

Patients were randomly assigned to amitryptyllin (AT), gabapentin (GB), pregabalin (PG), or placebo. AT was given at 50 mg/day for one week, then increased incrementally to 100 mg/day. GB was given at 900 mg/day for one week in divided doses, then increased to 1,800 mg/day by week 3. PG was given at 150 mg/day for one week, then increased to 600 mg/day by week 3. The control group received placebo capsules. Morphine was used as rescue pain medication as needed. Patients were evaluated weekly for four weeks.

• The study reported on 120 patients.
• Information on patient age and key disease characteristics was not provided.

• Single site    
• Setting not specified
• India

The study has clinical applicability for late effects and survivorship.

The study was a placebo-controlled, randomized trial.

• Visual analog scale (VAS) to measure pain globally as well as burning, lancinating pain and dysesthesia
• Eastern Cooperative Oncology Group (ECOG) score for functional capacity
• Global satisfaction rating on a five-point verbal scale

VAS scores decreased in all groups. In week 4, those on pregabalin had significantly lower VAS scores than other groups, and scores declined by 4–5 points (p < 0.03). By the third visit, the percentage of patients who required morphine rescue increased: 46.7% with AT, 23.3% with GB, 16.7% with PG, and 100% with placebo. Lancinating pain incidence was lowest in the PG group. There were significantly fewer patients with dysesthesia in the PG group (6.7%) compared to the GB and placebo groups after four weeks. The percentage of patients with allodynia declined in all groups. The PG group showed a statistically significant improvement in ECOG score compared to all other groups (p < 0.001). Satisfaction was similar in all groups, and there were no significant differences in adverse reactions, with a gradual increase in all groups over time. Adverse effects were somnolence, dizziness, nausea, constipation, and dry mouth.

All of the drugs tested here demonstrated some efficacy in improving neuropathic symptoms. In several areas, it appears that pregabalin was more effective than gabapentin, opioid monotherapy, and amitryptylline. All medications were given in combination with opioids for pain management.

• No information of opioid dosages in each group was provided for comparison.
• The follow-up period of four weeks was relatively short, given the usual duration of symptoms.
• No information was provided regarding the causes of peripheral symptoms (whether they were chemotherapy-induced or related to other causes).

All of the medications examined in this study were effective and had a morphine-sparing effect in the treatment of neuropathic pain and other symptoms. Pregabalin was more effective than other alternatives tested in some areas. As all patients in this study received opioids for pain rescue, it should be noted that essentially all medications compared were given in combination with opioids.