Investigating the effects of olanzapine on appetite and weight in patients with cancer in a systematic review

Abstract

Purpose

Anorexia, malnutrition, and cachexia are common in patients with cancer, yet evidence-based pharmacological guidance remains limited. Olanzapine is an antipsychotic that can increase appetite and cause weight gain. This systematic review evaluated the effects of olanzapine on appetite and body weight in patients with cancer.

Methods

This systematic review included randomized controlled trials (RCTs) and observational studies assessing the effects of olanzapine on appetite and/or body weight in patients with cancer. The search strategy was conducted in Medline (PubMed), Embase, and the Cochrane Library. Risk of bias was assessed using Cochrane RoB2 for RCTs and the ROBINS-I for observational studies.

Results

Of 1,733 records identified, 79 reports underwent full-text review. Fourteen RCTs, one prospective observational study, and three retrospective studies met the inclusion criteria. Most studies were not designed primarily to evaluate olanzapine as an orexigenic intervention; instead, olanzapine was typically studied for other indications (e.g. chemotherapy-induced nausea and vomiting control), with appetite and/or weight reported as secondary outcomes. Weight gain with olanzapine was reported in two RCTs using 2.5-5 mg/day for 8–12 weeks in adults. Increased appetite was reported in 10 of 14 studies in which olanzapine (2.5–10 mg/day) was administered for 3 days to 12 weeks. However, only three studies prospectively assessed appetite as a primary outcome, and only two prespecified body weight change as an endpoint, indicating that most evidence derives from secondary analyses. Most RCTs raised some concerns or were at high risk of bias, and observational studies generally had serious or critical risk of bias for outcomes of interest. Only 2 RCTs were judged to be at low risk of bias, underscoring substantial methodological limitations and heterogeneity in populations, indications, dosing regimens, and outcome definitions.

Conclusions

Available evidence suggests that olanzapine may have orexigenic effects in patients with cancer, with increased appetite reported in several studies and body weight gain observed with long-term use. These findings should be interpreted cautiously given the heterogeneity in study designs and the generally limited methodological quality of the evidence. Further research is needed to confirm these effects and to define the optimal dose and treatment duration for improving appetite and body weight in patients with cancer.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12672-026-05346-w.

Introduction

Patients with cancer frequently experience anorexia, malnutrition, and cachexia which represent a continuum of interrelated metabolic and nutritional disorders arising from a complex interplay among the malignancy itself, anticancer treatments, and cancer-related metabolic dysregulation [1–6]. These conditions are associated with increased morbidity and mortality, reduced efficacy of anticancer treatments [7–9] and survival [10–12], higher risks of post-operative complications [13], treatment-related toxicity [14, 15], and infections [16], as well as longer hospital stays and higher costs [13, 16, 17]. They also substantially impair quality of life (QoL) [18–21].

Despite this considerable clinical burden, effective pharmacological options for cancer-related anorexia, malnutrition, and cachexia remain limited, and no drug has been formally approved specifically for these indications. Recently, however, clinical practice guidelines have begun to consider potential pharmacological approaches. The 2021 ESMO clinical practice guidelines and the 2023 ASCO guidelines update conditionally recommend low-dose olanzapine for management of cancer cachexia [22, 23]. The inclusion of olanzapine in these guidelines reflects a drug-repurposing rationale, as olanzapine is well known to increase appetite and cause weight gain in patients treated for psychiatric disorders [24, 25].

Important uncertainties nevertheless remain regarding its role in oncology. Although current recommendations focus on low-dose olanzapine, the magnitude of its orexigenic effect, its potential impact across the spectrum of anorexia, malnutrition and cachexia, and the optimal dose, duration and patient selection have not been clearly established. Moreover, the available evidence is fragmented and has not been comprehensively synthesized.

Accordingly, this systematic review aims to address this gap by systematically synthesizing and critically appraising the evidence on the effects of olanzapine on appetite- and body weight-related outcomes in patients with cancer.

Methods

This systematic review was reported in accordance with the Preferred Reporting Items in Systematic Reviews and Meta-Analyses (PRISMA) statement [26] (PRISMA Checklist available in Appendix A). The protocol was registered in PROSPERO (CRD42024500191).

Search strategy

On September 9, 2024, we searched MEDLINE (via PubMed), Embase, and the Cochrane Library using the terms related to anorexia, malnutrition, cachexia, cancer, and olanzapine (full search strategy available in Appendix B). We also searched Google Scholar, ClinicalTrials.gov, and the WHO International Clinical Trial Registry Platform (ICTRP).

Eligibility criteria

We included fully published randomized controlled trials (RCTs) and observational studies (prospective or retrospective) involving adults or pediatric patients with cancer. We included both RCTs and observational studies to provide a comprehensive overview in an emerging area with limited studies data on olanzapine’s effects on appetite and body weight in oncology. Eligible interventions included olanzapine administered for orexigenic or non-orexigenic indications, compared with placebo, standard of care, or other orexigenic treatments, irrespective of treatment duration. Studies were required to report change in appetite and/or body weight. We excluded non-primary studies and studies limited to radiotherapy-only populations, as well as studies in non-cancer patients, anorexia nervosa, animals, or in vitro models. We also excluded reports without an appropriate comparator (when applicable), without relevant appetite/body weight outcomes, duplicate publications, and studies without accessible full text. No restrictions were applied regarding publication date or language.

Study selection and data extraction

Two reviewers (MCRT and KS) independently screened titles/abstracts and full-text articles. Disagreements were resolved by a third reviewer (FS).

MCRT and KS extracted study characteristics, methodological features, population details, interventions/comparators, and outcomes; discrepancies were resolved by discussion. The primary outcomes were changes in appetite (assessed using study-appropriate instrument/scales) and changes in body weight (kg). Secondary outcomes included change in energy intake (kcalories), QoL (excluding instruments specific to nausea/vomiting-related QoL, such as the Functional Living Index-Emesis [FLIE]), and olanzapine-related adverse events.

To enhance conceptual clarity, we distinguished between (1) appetite-related outcomes (e.g. self-reported appetite, desire to eat, or intake as defined by the original study) and (2) weight-related outcomes (e.g., body weight or body composition). Terms such as malnutrition and cachexia were used only when explicitly reported in the included studies.

Data synthesis

Given substantial heterogeneity in populations, indications, interventions/comparators, follow-up durations, outcome definitions, and measurement tools, meta-analysis was not performed. Heterogeneity was considered at three levels: clinical (e.g., cancer type, disease stage, setting), methodological (e.g., design, sample size, follow-up, risk of bias); and outcome-related (e.g., definitions and instruments for appetite/weight endpoints).

Although a random-effects meta-analysis and exploratory subgroup analyses were initially considered, they were not feasible due to the limited number of clinically comparable studies and the extent of heterogeneity. Appetite- and body weight-related endpoints were also frequently reported as secondary outcomes, further limiting comparability. Accordingly, we conducted a structured narrative synthesis. Studies were grouped by outcome type (appetite-related versus weight-related) and by whether the outcome was pre-specified as primary or secondary. No data transformation, sensitivity analyses, formal assessment of reporting bias or certainty assessment were performed due to the narrative nature of the synthesis and absence of meta-analysis.

Within groups, findings were compared with consideration of study design, sample size, follow-up, and the directness of outcome measurement. Evidence derived from antiemetic-focused studies was explicitly distinguished from studies designed to directly assess appetite or body weight. Reporting followed the Synthesis Without Meta-analysis (SWiM) guidelines [27]. Results are presented separately for RCTs and observational studies.

Risk of bias assessment

Two reviewers (MCRT and SM or PN) independently assessed risk of bias using RoB2 for randomized trials [28] (including the crossover extension when applicable) and ROBINS-I for observational studies [29]. Disagreements were resolved through discussion, with adjudication by a third reviewer (CC) when needed.

Results

As shown in the study flowchart (Fig. 1), the search identified 1,733 records. After screening, 79 reports were retrieved for full-text review. Eighteen studies met the eligibility criteria: 14 RCTs [30–43], one prospective observational study [44], and three retrospective studies [45–47], comprising 1,754 participants in total. Details of the excluded studies are provided in Appendix C.

Fig. 1.

Study selection

Table 1 provides an overview of the main study characteristics, including country, study design, sample size, cancer type, anticancer regimen, study outcomes, details of the intervention, and characteristics of the comparator group. All but one study enrolled adults; one study included 31 pediatric patients [43]. In most studies, participants were receiving chemotherapy regimens appropriate to their cancer type; in two studies [36, 39], patients were not receiving chemotherapy at the time of enrollment.

Table 1.

Baseline characteristics of included studies

References	Country (HDI)	Study design	No. of groups& No. pts analyzed per group	Cancer	Anticancer regimen	Study outcomes	Intervention	Comparator
Sandhya et al. [38]	India (0.644)	RCT, Monocentric	Two groups of adults Intervention: 58 Comparator: 54	Gastric, lung, Hepatopancreatico-biliary Stage 3 or 4	CAPOX, FLOT, DOX, GEMOX/GemCap, Paclitaxel and carboplatin, pemetrexed and carboplatin, pemetrexed and cisplatin	Primary outcome: Appetite, weight Secondary outcomes: Energy intake, QoL, AE	Olanzapine 2.5 mg/day 12 weeks	Placebo 12 weeks
Navari et al. [39]	USA (0.927)	RCT, Monocentric	Two groups of adults Intervention: 39 Comparator: 37	Colon, lung Stage 3 or 4	No treatment	Primary outcome: Appetite, Weight Secondary outcomes: QoL, AE	Olanzapine 5 mg/day + Megestrol acetate 800 mg/day 8 weeks	Megestrol acetate 800 mg/day 8 weeks
Okamoto et al. [46]	Japan (0.920)	Observational retrospective, Monocentric	One group of adults Before intervention: 80 After intervention: 80	Digestive organs, gynecological organs, respiratory, nephro-urinary organs, blood, head and neck, others	NR	Primary outcome: Food intake Secondary outcomes: AE	After Olanzapine 2.28 ± 0.87 mg/day 3 days	Before olanzapine
Yeo et al. [30]	China (0.788)	RCT, Monocentric	Two groups of women Intervention: 60 Comparator: 60	Breast cancer Stage 1, 2 or 3	AC regimen	Primary outcome: CINV Secondary outcomes: Weight, Anorexia, AE	Olanzapine 10 mg/day 5 days + Aprepitant 125 mg day1, 80 mg days 2–3 + Ondansetron 8 mg bid day1 + Dexamethasone 12 mg day1	Aprepitant 125 mg day1, 80 mg days 2–3 + Ondansetron 8 mg bid day1 +Dexamethasone 12 mg day1, 4 mg bid days 2–3
Wu et al. [47]	China (0.788)	Observational retrospective, Multicentric	Two groups of women Intervention: 40 Comparator: 53	Colorectal, gastric	Oxaliplatin containing or irinotecan containing regimens	Primary outcome: CINV Secondary outcomes: Weight, Anorexia, AE	Olanzapine 5 mg/day 3 days + Dexamethasone 12 mg days 1–3 + Tropisetron 5 mg day1	Dexamethasone 12 mg days 1–3 + Tropisetron 5 mg day1
Clemons et al. [31]	Canada (0.935)	RCT, multicentric	Two groups of women Intervention: 113 Comparator: 105	Breast cancer Stage 1, 2 or 3	AC, FEC or TCH chemotherapy regimens	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 5 mg/day 4 days + Aprepitant 125 mg day 1, 80 mg days 2–3 + Ondansetron 8 mg bid day1 + Dexamethasone 12 mg day1, 4 mg bid days 2–3	Aprepitant 125 mg day 1, 80 mg days 2–3 + Ondansetron 8 mg bid day1 + Dexamethasone 12 mg day1, 4 mg bid days 2–3
Vimolchalao et al. [32]	Thailand (0.803)	Crossover RCT, Monocentric	Two groups of adults 1st group: 30 2nd group: 30	Breast, head and neck, lung, other	AC or high-dose cisplatin chemotherapy regimen (100%), radiotherapy (34.38%)	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 10 mg/day 4 days + Ondansetron 8 mg day1 + Dexamethasone 20 mg day1, 4 mg bid days 2–4	Matching placebo + Ondansetron 8 mg day1 + Dexamethasone 20 mg day1, 4 mg bid days 2–4
An et al. [34]	China (0.788)	RCT, Monocentric	Two groups of adults Intervention: 56 Comparator: 57	Lung, other	Combined carboplatin chemotherapy regimen	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 5 mg/day 4 days + Tropisetron 5 mg day1 + Dexamethasone 10 mg day1	Tropisetron 5 mg day1 + Dexamethasone 10 mg day1
Mizukami et al. [35]	Japan (0.920)	RCT, Multicentric	Two groups of adults Intervention: 22 Comparator: 22	Bladder, lymphoma, pharynx, breast, leukemia, others	Cisplatin, EC, AC, nedaplatin, carboplatin, daunorubicin or others chemotherapy regimens	Primary outcome: CINV Secondary outcomes: Dietary intake, AE	Olanzapine 5 mg/day 6 days + Dexamethasone 9.9 mg day1, 6.6 mg days 2–4 + Granisetron 3–6 mg days1-3 or ondansetron 4 mg days 1–2 or ramosetron 0.6 mg days 1–3 or palonosetron 0.75 mg day 1 + Aprepitant 125 mg aprepitant day 1 and 80 mg days 2–3	Placebo + Dexamethasone 9.9 mg day1, 6.6 mg days 2–4 + Granisetron 3–6 mg days1-3 or ondansetron 4 mg days 1–2 or ramosetron 0.6 mg days 1–3 or palonosetron 0.75 mg day 1 + Aprepitant 125 mg aprepitant day 1 and 80 mg days 2–3
Navari et al. [36]	USA (0.927)	RCT, Multicentric	Two groups of adults intervention: 15 Comparator: 15	Breast, colon, gastric, gynecologic, head & neck, pancreatic, prostate, lung, genitourinary Previously treated	No treatment	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 5 mg/day 7 days	Matched placebo
Liu et al. [37]	China (0.788)	RCT, Monocentric	Two groups of adults intervention: 108 Comparator: 106	Lung, breast, colorectal, stomach, lymphoma, ovarian, oesophageal, teratoma, oropharyngeal, thymus, cervical, gingival, malignant melanoma, laryngeal, glioblastoma	Cisplatin, oxaliplatin, epirubicin, carboplatin, dacarbazine or doxorubicin-based regimen	Primary outcome: CINV Secondary outcomes: Appetite, QoL	Olanzapine 10 mg/day 5 days + Azasetron 10 mg day1 + Dexamethasone 10 mg day1	Azasetron 10 mg day1 + Dexamethasone 10 mg day1
Mehra et al. [44]	India (0.644)	Observational prospective, Monocentric	one group of adults Before intervention: 54 After intervention: 54	Breast, Sarcoma, Head & Neck, germ cell tumor, lung, cervix, bladder	AC, FEC, cisplatin or doxorubicin regimen	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 10 mg/day 4 days + Palonosetron 0.25 mg day1 + Dexamethasone 12 mg day1	Aprepitant 125 mg day1, 80 mg days 2–3 + Palonosetron 0.25 mg day1 + Dexamethasone 12 mg day1
Suehiro et al. [45]	Japan (0.920)	Observational retrospective, Monocentric	Two groups of women Intervention: 82 Comparator: 95	Breast cancer Stage 1, 2 or 3	FEC, EC, dose-dense EC regimen	Primary outcome: CINV Secondary outcomes: Anorexia, AE	Olanzapine 5 mg/day 4 days + Aprepitant 125 mg day 1 and 80 mg days 2–3 + Palonosetron 0.75 mg day1 + Dexamethasone 12 mg for EC and FEC, 10 mg for dose-dense EC, day 1	Aprepitant 125 mg day 1 and 80 mg days 2–3 + Palonosetron 0.75 mg day1 + Dexamethasone 12 mg for EC and FEC, 10 mg for dose-dense EC, day 1
Nikbakhsh et al. [33]	Iran (0.780)	RCT, Monocentric	Two groups of adults Intervention: 15 Comparator: 15	Gastric	NR	Primary outcome: QoL Secondary outcomes: Appetite (based on BMI)	Olanzapine 2.5–10 mg/day 8 weeks and 2 days + “Routine treatment regimens”	“Routine treatment regimens”
Maleki et al. [40]	Iran (0.780)	RCT, Multicentric	Two groups of women Intervention: 30 Comparator: 30	Breast cancer Stage 1, 2 or 3	AC regimen	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 10 mg or 5 mg if ≥ 60 years/day 4 days + Aprepitant 125 mg day1 and 80 mg days 2–3 + Dexamethasone 12 mg day1 + Granisetron 1 mg day1	Mirtazapine 15 mg/day 4 days + Aprepitant 125 mg day1 and 80 mg days 2–3 + Dexamethasone 12 mg day1 + Granisetron 1 mg day1
Shen et al. [41]	China (0.788)	RCT, Monocentric	Two groups of women Intervention: 51 Comparator: 51	Gastric, colon or rectal	XELOX, SOX, FOLFOX, FOLFIRI or XELIRI regimen	Primary outcome: CINV Secondary outcomes: Appetite, AE	Olanzapine 5 mg/day 5 days +Dexamethasone 10 mg day1 +Palonosetron 0.25 mg day1	Dexamethasone 10 mg day1 + Palonosetron 0.25 mg day1
Ithimakin et al. [42]	Thailand (0.803)	RCT, Monocentric	Two groups of adults Intervention: - OLZ 5 mg: 47 - OLZ 10 mg: 46 Comparator: 47	Breast, head and neck, lung, others	AC/cyclophosphamide or cisplatin ≥ 50mg/m2 based regimen	Primary outcome: CINV Secondary outcomes: Anorexia, AE	Olanzapine 5 mg or 10 mg/day 4 days + Ondansetron 8 mg day1 + Dexamethasone 12 mg day1 and 8 mg days 2–4 + Aprepitant matched placebo	Olanzapine matched placebo + Aprepitant 125 mg day1, 80 mg days 2–3 + Ondansetron 8 mg day1 + Dexamethasone 12 mg, day 1 and 8 mg days 2–4
Zhukov et al. [43]	Russia (0.822)	Crossover RCT, Monocentric	Two groups of pediatric pts Group A: 17 Group B: 14	Ewing sarcoma, osteosarcoma, other sarcomas, mesenchymal and embryonal tumors, nasopharyngeal tumor, lymphoepithelioma-like tumor, germ cell tumor	VDC, doxorubicin + cisplatin, PEI, carboplatin + epirubicin + vincristine, 5-fluorouracil + cisplatin, carboplatin + etoposide, vincristine + Ifosfamide + doxorubicin, ifosfamide + etoposide, vincristine + ifosfamide + etoposide, methotrexate, N6, cisplatin + gemcitabin, cyclophosphamide + doxorubicin	Primary outcome: CINV Secondary outcomes: Weight, AE	Olanzapine 2.5-5 mg/day During day(s) of CT + 3 days after completion + Dexamethasone 5 mg/m2 + Ondansetron 0,15 mg/kg every 8 h during day(s) of CT and 3 days after CT + Patients weighing 30–40 kg: Aprepitant 80 mg once during day(s) of CT + 2 days after CT + Patients > 40 kg: Aprepitant 125 mg during day(s) of chemo and 80 mg 2 days after CT	Dexamethasone 5 mg/m2 + Ondansetron 0,15 mg/kg every 8 h during day(s) of CT and 3 days after CT + Patients weighing 30–40 kg: Aprepitant 80 mg once during day(s) of CT + 2 days after CT + Patients > 40 kg: Aprepitant 125 mg during day(s) of chemo and 80 mg 2 days after CT

AC: Adriamycin/Doxorubicin and cyclophosphamide; AE: Adverse Events; BMI: Body Mass Index; CAPOX: capecitabine and oxaliplatin; CDDP: Cisplatin; CINV: Chemotherapy-induced nausea and vomiting; CT: chemotherapy; DOX: docetaxel, oxaliplatin, capecitabine; EC: Epirubicin and cyclophosphamide; FEC: 5-Fluorouracil, epirubicin, cyclophosphamide; FLOT: 5-Fluorouracil, oxaliplatin, docetaxel; FOLFIRI: folinic acid, 5-Fluorouracil, irinotecan; FOLFOX: folinic acid, 5-Fluorouracil, oxaliplatin; GEMOX: gemcitabine and oxaliplatin; GemCap: gemcitabine and capecitabine; NA: Not Applicable; NK-1 RA: NK-1 Receptor Antagonist; NSCLC: Non-Small Cell Lung Cancer; No.: Number; N6: vincristine, dacarbazine, ifosfamide, doxorubicin; PEI: etoposide, cisplatin, ifosfamide; Pts : patients; QoL : Quality of Life; SOX: S-1, oxaliplatin ; TCH: Docetaxel, cyclophosphamide, trastuzumab; TKI: Tyrosine kinase inhibitors; VDC: vincristine, doxorubicin, cyclophosphamide; XELIRI: capecitabin, irinotecan; XELOX: capecitabin, oxaliplatin; 5-HT3 RA: 5-HT3 Receptor Antagonist

This review synthesizes both direct evidence (studies in which appetite or weight outcomes were prespecified primary endpoints) and indirect evidence (studies in which these outcomes were secondary, often in antiemetic trials). These evidence types are not equivalent and differ in interpretability. Outcomes measures for appetite and body weight were heterogenous across studies, limiting comparability.

Body weight-related outcomes

All body weight-related outcomes are summarized in Table 2.

Table 2.

Body weight related outcomes

Reference		Sandhya et al. [38]			Navari et al. [39]			Zhukov et al. [43]		Yeo et al. [30]	Wu et al. [47]	Nikbakhsh et al. [33]
Study design		RCT			RCT			Crossover RCT		RCT	Retrospective	RCT
Population		112 adults			76 adults			31 pediatric patients		120 adults (women)	93 adults	30 adults
Outcomes		> 5% Weight gain (pts, no.; %)	Weight loss (%pts)	Mean weight (± SD)	≥ 5% Weight gain (pts, no.; %)			Weight gain > 1000 g (pts, no; %)	Weight loss > 1000 g (pts, no; %)	Median weight (IQR)	Weight loss by 5% (pts, no.; %)	Appetite based on BMI (mean (SD))
Time of measurement		12 weeks			4&8 weeks	4 weeks	8 weeks	After 1 cycle of chemotherapy		Day 5	After 1st course of chemotherapy	8th week
Results	Olanzapine group	35/58; 60%	14%	55.7 ± 9.05 kg baseline: 53.1 ± 8.6 kg p < 0.001	33/39; 85%	35/39; 90%	33/39; 85%	Std first: 8/14; 54% OLZ first:13/17;76%	Std first: 0/14; 0% OLZ first: 1/17 6%	57.8 kg (42.3–81.2) baseline: 57.3 kg (41.6–82.7)	4/40; 10%	23.30 ± 2.99 (baseline: 22.76 ± 3.21)
	Comparator	5/54; 9%	59%	51.7 ± 9.7 kg baseline: 53.6 ± 9.9 kg p < 0.001	15/37; 41%	21/37; 57%	15/37; 41%	Std first: 0/14; 0% OLZ first: 0/17; 0%	Std first: 10/14; 69% OLZ first: 11/17; 65%	57.8 kg (46.5–105.0) baseline: 58.9 kg (44.5-100.4)	4/53; 7.5%	21.38 ± 4.21 (baseline: 21.66 ± 3.78)
	Groups difference	p<0.0001	NR	NR	NR	NR	NR	Std first: p = 0.023 OLZ first: p < 0.001	Std first: p = 0.008 OLZ first: p = 0.009	p = 0.4326 baseline: p = 0.2184	p = 0.722	p = 0.16

BMI: Body Mass Index; NR: Not Reported; OLZ: Olanzapine; pts: patients; RCT: Randomized Controlled Trial; Std: Standard treatment

Studies assessing changes in body weight as a primary outcome

In the RCT by Sandhya et al. [38], patients with gastrointestinal or lung cancer receiving olanzapine 2.5 mg/day for 12 weeks were more likely to achieve > 5% weight gain than those receiving placebo (between-group difference, p > 0.0001). Mean body weight increased from baseline to week 12 in the olanzapine group (p < 0.001) whereas it decreased in the placebo group over the same period (p < 0.001). The proportion of patients with weight loss at week 12 was lower with olanzapine than with placebo (14% vs. 59%).

Navari et al. [39] conducted a RCT in patients with advanced colon or lung cancer who had not received chemotherapy within 4 weeks before enrollment. A ≥ 5% weight gain at weeks 4 and 8 was reported in the olanzapine (5 mg/day) plus megestrol acetate compared with megestrol acetate alone.

Studies assessing body weight changes as a secondary outcome

In the RCT by Yeo et al. [30], olanzapine (10 mg once daily for 5 days) added to standard triplet antiemetic therapy (aprepitant, ondansetron, and dexamethasone) was not associated with a significant difference in median body weight vs. control (p = 0.4326).

In the RCT by Nikbakhsh et al. [33], olanzapine (2.5–10 mg/day for 8 weeks and 2 days) was associated with a within group increase in BMI in the intervention group (23.30 ± 2.99 kg/m², p = 0.001), whereas BMI did not change in the control group (21.38 ± 4.21 kg/m², p = 0.16). However, between-group differences were not statistically significant at baseline or at end of treatment.

In a retrospective study, Wu et al. [47] reported no difference in the proportion of patients with ≥ 5% weight loss between those receiving olanzapine (5 mg/day for 3 days) plus dexamethasone and tropisetron vs. dexamethasone and tropisetron alone (p = 0.722).

In a pediatric crossover RCT [43], patients received olanzapine (2.5–5 mg/day, weight-adjusted) plus triplet antiemetic therapy (dexamethasone, aprepitant, and ondansetron), or the same regimen without olanzapine, across consecutive chemotherapy cycles. Weight gain (> 1000 g) was observed during cycles including olanzapine, whereas weight loss (> 1000 g) was observed during cycles without olanzapine.

Appetite-related outcomes

All the appetite-related outcomes are summarized in Tables 3 and 4.

Table 3.

Appetite-related outcomes

Reference		Sandhya et al. [38]				Navari et al. [39]						Clemons et al. [31]	Navari et al. [36]	Vimolchalao et al. [32]	An et al. [34]	Maleki et al. [40]
Study design		RCT				RCT						RCT	RCT	Crossover RCT	RCT	RCT
Population		112 adults				76 adults						218 adults	30 adults	60 adults	113 adults	60 adults (women)
Type of outcome		Appetite improvement (pts No.; %)	FAACT ACS score > 37 (pts No.; %)	Appetite score (median (range))	FAACT ACS score (median (range))	Appetite improvement (pts No.; %)			MDASI appetite score (mean (SD))			Increased appetite (% by cycle)	Changes in intensity of appetite (median (range))	Appetite (mean)	Loss of appetite (pts No.; %)	Loss of appetite (pts No.; %)
Measurement tool		VAS ≥ 3-points	FAACT ACS v2.0	VAS (score 1–10)	FAACT ACS v2.0	VAS + 3 points			MDASI (score 0–10)			CTCAE v4.02	Numeric rating score (0–10)	VAS (score 0–10)	CTCAE v4.0	CTCAE v5.0
Time of measurement		12 weeks				4&8 weeks	4 weeks	8 weeks	0 weeks	4 weeks	8 weeks	NR	From baseline to day 7	Days 1–5	Day 21	Cycle 1 Days 1–5	Cycle 2 Days 1–5
Results	Olanzapine group	25/58; 43%	13/58; 22%	6 (2–9) baseline: 4 (1–7)	33.5 (11–48) baseline: 17 (4–28)	25/39; 64%	27/39; 69%	25/39; 64%	1.5 (1.9)	5.7 (4.1)	6.3 (3.7)	12.8% of 383 cycles	5 (4 to 7)	2.5	28/56; 50.00%	Grade 1 1/30; 3.3% Grade 2 10/30; 33.3%	Grade 1 4/29; 14.3% Grade 2 9/29; 32.1%
	Comparator	7/54; 13%	2/54; 4%	3 (1–8) baseline: 4 (1–7)	20 (7–39) baseline: 14 (7–37)	2/37; 5%	5/37; 14%	2/37; 5%	1.7 (2.1)	2.1 (2.7)	2.0 (2.3)	6.6% of 346 cycles	0 (-1 to 2)	1.55	40/57; 70.17%	Grade 1 1/30; 3.3% Grade 2 12/30; 40.0%	Grade 1 4/29; 13.8% Grade 2 7/29; 24.1%
	Groups difference	p < 0.001	p = 0.004	NR	NR	NR	NR	NR	NR	NR	NR	NR	5 (5–6) p < 0.001	p = 0.003	p = 0.028	p = 0.59	p = 0.49

CTCAE: Common Terminology Criteria for Adverse Events; FAACT ACS scores: Functional Assessment of Chronic Illness Therapy system of Quality-of-Life questionnaires Anorexia Cachexia subscale; MDASI: MD Anderson Symptom Inventory; NR: Not Reported; pts: patients; RCT: Randomized Controlled Trial; VAS: Visual Analog Scale

Table 4.

Appetite-related outcomes

Reference		Shen et al. [41]		Liu et al. [37]	Mehra et al. [44]	Yeo et al. [30]		Suehiro et al. [45]	Wu et al. [47]	Ithimakin et al. [42]	Mizukami et al. [35]	Okamoto et al. [46]
Study design		RCT		RCT	Prospective observational	RCT		Retrospective	Retrospective	RCT	RCT	Retrospective
Population		102 adults		214 adults	54 adults	120 adults (women)		177 adults	93 adults	140 adults	44 adults	80 adults
Type of outcome		Loss of appetite (pts No.; %)	Appetite enhancement (pts No.; %)	Appetite reduction (mean score change)	Lack of appetite (median score (range))	Anorexia (pts No.; %)		Anorexia (pts No.; %)	Anorexia (pts No.; %)	Anorexia (pts No.; %)	Amount in dietary intake compared with day 1 (graphic)	Relative change in food intake
Measurement tool		NR	NR	EORTC QLQ C30 V3.0 Chinese version score (0-100)	MDASI score (0–10)	NCI CTCAE v4.0		CTCAE v4.0	VAS score	NR	NR	Nurse measurements
Time of measurement		Days 2 to 6		Before/after chemotherapy	Day 1–5	Day 5		NR	After 1st course of chemotherapy	Days 2–5	Days 1–6	Days 1–3
Results	Olanzapine group	29/51; 56.86%	22/51; 43.14%	-4.33	2 (0–10)	Grade 0–1: 60/60; 100%	Grade 2: 0/60; 0%	Grade 0: 53/82; 64.6% Grade 1: 14/82; 17.1% Grade 2: 15/82; 18.3% Grade 3: 0/82; 0%	23/40; 57.5%	OLZ 10 mg: 13/46; 29.5% OLZ 5 mg: 19/47; 44.2%	Maintain the amount of dietary intake	Comparison before and after olanzapine administration: 149% on average p < 0.0001
	Comparator	37/51; 72.55%	14/51; 27.45%	39.70	8 (0–10)	Grade 0–1: 57/60; 95%	Grade 2: 3/60; 5%	Grade 0: 21/95; 22.1% Grade 1: 48/95; 50.5% Grade 2: 26/95; 27.4% Grade 3: 0/95; 0%	42/53; 79.2%	19/47; 46.3%	Significantly reduced on days 5 and 6
	Groups difference	p = 0.097	p = 0.036	p = 0.000	p < 0.001	p = 0.1218		NR	p = 0.024	OLZ 10 mg: p = 0.17 OLZ 5 mg: p = 0.97	Days 1 & 4: p < 0.05 Days 5 & 6: p < 0.01

CTCAE: Common Terminology Criteria for Adverse Events; EORTC QLQ-C30: European Organization for Research and Treatment quality of life (QOL) questionnaire; MDASI: MD Anderson Symptom Inventory; NR: Not Reported; OLZ: Olanzapine; pts: patients; RCT: Randomized Controlled Trial; VAS: Visual Analog Scale

Studies assessing appetite as a primary outcome

In the RCT conducted by Sandhya et al. [38], olanzapine 2.5 mg/day daily significantly improved appetite from baseline to week 12 compared with placebo (p < 0.001). This RCT [38] also reported higher FAACT A/CS (Functional Assessment of Chronic Illness Therapy system of Quality-of-Life questionnaires Anorexia Cachexia subscale) scores with olanzapine versus placebo (p = 0.004). Using a 24-hour dietary recall, 52% of patients in the olanzapine group (30/58) met 75% of their calorie and protein needs, compared with 18% in the placebo group (10/54) (p < 0.0001). Nutritional status, assessed using the Subjective Global Assessment (SGA), improved more frequently with olanzapine (43%, 25/58) than with placebo (9%, 5/54) (p < 0.0001).

Another RCT [39] reported greater appetite improvement at weeks 4 and 8 with olanzapine 5 mg/day plus megestrol acetate compared with megestrol acetate alone.

In a retrospective study, Okamoto et al. [46] found that olanzapine (mean dose 2.28 mg/day) was associated with a 149% increase in food intake after 3 days of treatment (p < 0.0001).

Studies assessing appetite as a secondary outcome

In the RCT by Navari et al. [36], patients received olanzapine 5 mg/day or placebo for 7 days. Appetite intensity, assessed daily using a 0–10 numeric rating scale, was higher in the olanzapine group than in the placebo group (5 vs. 0; p < 0.001).

In the RCT by Shen et al. [41] olanzapine 5 mg/day for 5 days combined with dexamethasone and palonosetron was associated with improved appetite compared with control group (43.14% vs. 27.45%, p = 0.036). Similarly, Liu et al. [37] also reported improved appetite with olanzapine 10 mg/day for 5 days added to azasetron and dexamethasone (p = 0.000) in a RCT.

In the RCT by Clemons et al. [31], an increase in appetite was observed in 12.8% vs. 6.6% of chemotherapy cycles with the addition of 5 mg olanzapine daily for 4 days to standard antiemetics compared with standard antiemetics alone.

In the RCT by An et al. [34], olanzapine 5 mg/day for 4 days (with tropisetron and dexamethasone) reduced appetite loss compared with control (50% vs. 70%, p = 0.028).

In the RCT by Maleki et al. [40], no differences in appetite loss were observed between olanzapine 5–10 mg/day and mirtazapine 15 mg (both for 4 days) during the first two chemotherapy cycles (p = 0.59 and p = 0.49).

In a crossover RCT, Vimolchalao et al. [32], reported higher appetite visual analog scale (VAS) scores with olanzapine 10 mg/day for 4 days plus ondansetron and dexamethasone compared with placebo plus the same antiemetic regimen (p = 0.003).

In the RCT by Yeo et al. [30], grade 0–1 anorexia occurred in 100% of patients receiving olanzapine (10 mg/day for 5 days) plus standard antiemetics (aprepitant, ondansetron and dexamethasone) and in 95% receiving standard antiemetics alone (p = 0.1218).

In a RCT, Ithimakin et al. [42] found no differences in anorexia among olanzapine 5 mg/day, olanzapine 10 mg/day, and aprepitant groups (p = 0.17 to p = 0.97).

In another RCT, Mizukami et al. [35], reported that olanzapine 5 mg/day for 6 days, added to standard antiemetics, increased dietary intake on days 2, 4, 5, and 6 (p < 0.05 to p < 0.01), with intake remaining stable in the olanzapine group while declining in the control group.

In a prospective observational study, Mehra et al. [44] reported less appetite loss in a within-group pre-post comparison with olanzapine 10 mg/day for 4 days (score 2/10 measured with MDASI (MD Anderson Symptom Inventory) lack of appetite 0 to 10 score) compared with aprepitant (8/10, p < 0.001).

In retrospective studies, Suehiro et al. [45] reported lower anorexia frequency in the olanzapine group than in controls (35.4% vs. 77.9%), and Wu et al. [47] reported lower anorexia rates with olanzapine (57.5% vs. 79.2%, p = 0.024).

Quality of life outcomes

In Sandhya et al. [38], QoL (Cancer Institute QoL questionnaire, version II) improved in 70% (44/58) of patients receiving olanzapine versus 50% (27/54) receiving placebo.

In Navari et al. [39], QoL measured using the Functional Assessment of Cancer Therapy-General (FACT-G v3) improved in 59% (23/39) of patients receiving olanzapine plus megestrol acetate versus 14% (5/37) receiving megestrol acetate alone.

In Nikbakhsh et al. [33], QoL measured using the WHO-QOL-BREF (Iranian version) improved significantly only in the physical health domain (p < 0.05), with no differences in other domains or overall QoL (p > 0.05).

In Liu et al. [37], QoL assessed with the EORTC QLQ-C30 v3.0 showed improvement in global health status in the olanzapine group (mean change 3.25) compared with control (-1.92; p = 0.005).

Adverse events

The most commonly reported adverse events were somnolence [34, 40, 45–47], sedation [31, 39, 43], drowsiness [35, 38, 41, 44], fatigue [40] or sleepiness [32, 42]. Adverse events reported in single studies included extrapyramidal symptoms [31], QTc prolongation [32], hyperbilirubinemia [38], constipation [38], hyperglycemia [38], headache [38] and dizziness [47]. Additional details are provided in Appendix D.

Risk of bias assessment

Randomized controlled trials: RoB2 tool

The 14 RCTs varied in methodological quality (Appendix E): two (14.3%) were judged at low risk of bias [36, 38], seven (50%) had some concerns [31, 32, 34, 39, 40, 42, 43], and five (35.7%) were at high risk [30, 33, 35, 37, 41]. Bias due to missing outcome data (D3) was low across studies. Bias arising from the randomization process (D1) was low in 11 studies (78.6%) [31–36, 38, 40–43], raised some concerns in one study (7.1%) [39], and was high in two studies (14.3%) [30, 37], primarily due to insufficient reporting on allocation concealment. Bias due to deviations from intended interventions (D2) was low in nine studies (64.3%) [30, 32–34, 36, 38, 39, 42, 43], raised some concerns in three [31, 37, 40], and was high in two (14.3%) [35, 41] due to limited information on blinding and appetite assessment methods. Bias in measurement of the outcome (D4) was low in seven studies (50%) [31, 32, 35, 36, 38, 40, 43], raised some concerns in four (28.6%) [34, 39, 41, 42], and was high in three (21.4%) [30, 33, 37], largely because outcome assessors were aware of group assignment, potentially influencing outcome measurement. Bias in selection of the reported result (D5) was low in seven studies (50%) [34, 36–40, 43] and raised some concerns in seven (50%) [30–33, 35, 41, 42]. The two crossover studies raised additional concerns regarding period and potential carryover effects because the washout duration was not reported [32, 43].

Non-randomized studies: ROBINS-I tool

Among observational studies (Appendix F), two retrospective studies were rated as having a serious risk of bias [45, 46], and the prospective study and one retrospective study were rated as having a critical risk of bias [44, 47]. Risk of bias was low for classification of interventions (D3), deviations from intended interventions (D4), and missing data (D5) across studies. Bias due to confounding (D1) was low in one study [45], serious in one [46], and critical in two studies [44, 47], reflecting limited control for key confounders and time-varying confounding. Participant selection bias (D2) was low in two studies [45, 46], and serious [47] or critical [44] in the remaining studies due to selection based on post-intervention characteristics. Bias in outcome measurement (D6) was moderate in two studies [46, 47] and serious in two [44, 45] because outcome assessment may have been influenced by awareness of intervention status. Bias in selection of the reported result (D7) was low in three studies (75%) [44, 45, 47] and moderate in one [46].

Discussion

Main findings

This systematic review evaluated the effects of olanzapine on appetite and body weight in patients with cancer. Although body weight gain is typically considered an adverse effect of olanzapine, it could be clinically beneficial in oncology, where anorexia and body weight loss substantially worsen QoL and prognosis.

We deliberately included studies in which olanzapine was prescribed for other indications in cancer care, particularly chemotherapy-induced nausea and vomiting (CINV), to capture a broader range of data on dose, duration, timing of effects on appetite and body weight, cancer types, and concurrent anticancer regimens. Trials in oncology have shown that olanzapine can reduce nausea and vomiting in patients receiving highly emetogenic regimens, which may indirectly improve appetite and facilitate weight gain.

However, this approach introduces an important interpretative challenge related to confounding by indication and symptom mediation. In many included studies, olanzapine was administered primarily for its antiemetic effects, making it difficult to separate a direct orexigenic effect from appetite improvement secondary to better control of nausea and vomiting. Consequently, observed benefits may be mediated, at least in part, by symptom relief rather than reflecting a direct pharmacologic effect on appetite regulation. Most studies were not designed to disentangle these mechanisms or to adjust for closely related symptoms.

Overall, olanzapine may improve appetite even with short-term use, potentially as a downstream consequence of improved nausea and vomiting control. In contrast, clinically meaningful body weight gain appears more evident with longer exposure, as suggested by two studies [38, 39]. This pattern is consistent with the possibility that appetite-related effects can emerge quickly, whereas sustained body weight changes may require prolonged treatment [48].

In this review, one study evaluating energy intake and two studies assessing dietary intake reported improvements with olanzapine group [35, 38, 46]. These findings are broadly consistent with observations in psychiatric populations, but additional studies are needed to confirm their magnitude and reproducibility in oncology.

Several studies also reported improvement in QoL with olanzapine; however, these findings should be interpreted cautiously given heterogeneous adverse effects reporting. Sedation-related effects (e.g., somnolence) were commonly reported and may be dose-dependent [49]. Although these adverse effects did not appear to substantially worsen QoL in the included studies, the absence of a formal benefit-risk assessment and inconsistencies in reporting limit firm conclusions regarding safety and overall benefit-risk balance. Olanzapine has also been associated with liver test abnormalities in approximately 10%-50% of treated patients, which are usually mild, transient, and may resolve spontaneously, potentially related to hepatic steatosis. Rarely, more marked aminotransferase elevations and hepatitis with jaundice have been reported. This safety profile warrants consideration in oncology, as hepatotoxicity could complicate continuation of anticancer therapy; importantly, liver test abnormalities typically improve after olanzapine discontinuation [50].

Potential mechanisms

Because appetite and nausea are closely linked, olanzapine may increase appetite through antagonism of 5-HT2C, H1, D2, and 5-HT2B receptors, potentially enhancing ghrelin signaling [51–53]. Low ghrelin levels have been associated with nausea, whereas weight gain appears more strongly related to H1 and 5-HT2C antagonism [53–56]. With chronic use, H1-mediated mechanisms may further increase food intake [57].

Future directions

Several included studies had methodological limitations that may affect reliability. For example, appetite was sometimes inferred from BMI change [33], although these are distinct constructs [58, 59], and body weight gain does not necessarily indicate improved lean body mass or nutritional status. Future trials should use validated appetite measures alongside objective endpoints such as weight, quantified intake, body composition and function, and selected biomarkers (e.g., albumin or transthyretin) [60]. Reporting of co-interventions (including rescue therapy for breakthrough CINV) should be standardized and complete. Given the higher risk of bias in observational studies, these findings should be considered hypothesis-generating. Indirect evidence from CINV-prevention trials can be informative but requires careful appraisal of relevance and internal validity; as highlighted by ASCO, higher-quality studies are needed for definitive conclusions [22]. Another systematic review also evaluated this topic, providing complementary insights into the subject [61]. Future research should consistently document cancer type and stage, anticancer regimen characteristics (type, duration, emetogenicity), and key baseline factors (age, sex, appetite, body weight), and should better reflect real-world populations experiencing anorexia, malnutrition and cachexia during active treatment. Dose allocation and comparator regimens should be clearly described and designs should account for olanzapine’s dual antiemetic and potential orexigenic roles through appropriate timing and follow-up. Analyses should report both within-group and between-group effects and include clinically meaningful endpoints, such as the proportion of patients achieving prespecified nutritional of functional targets.

Study limitations and strengths

A major limitation of our systematic review is that meta-analysis was not performed because of substantial heterogeneity in outcomes, interventions, comparators, populations, follow-up, and studies quality. Many results were indirect, and only one study reported energy intake; none directly evaluated malnutrition or cachexia using prespecified diagnostic criteria. Confounding by indication and variation in comparator regimens further complicate interpretation, and short follow-up in several studies limits assessment of longer-term effects. Moreover, most included studies focused primarily on olanzapine’s antiemetic effects properties in chemotherapy settings, which may not be fully generalizable to the broader population of cancer patients experiencing anorexia or cachexia. The diversity of instruments used to measure appetite and QoL also hindered cross-study comparisons. Potential bias stems from the inclusion of observational studies, which generally carry a higher risk of bias than RCTs.

External validity is limited, as is common in reviews pooling studies from diverse health systems and practice pattern. Across studies, most tumor types are represented (possibly excluding thyroid cancers), and anticancer regimens used alongside olanzapine were varied but not comprehensive. Variability in supportive care practices, including nonpharmacologic interventions, likely further limits generalizability.

Interpretation should also consider the overall risk of bias. Only 2 of 14 RCTs were judged at low risk of bias, 5 were at high risk, and observational studies were generally at serious to critical risk of bias, particularly due to confounding. Accordingly, positive signals should be interpreted cautiously.

Although formal assessment of publication bias was not feasible without meta-analysis and with few studies per outcome, reporting remains possible, as studies with positive findings may be more likely to be published.

This systematic review has several strengths. A comprehensive literature search was conducted across multiple databases without language restrictions, enhancing the inclusivity of evidence base. The review followed a structured methodological approach, with protocol registration and systematic risk of bias assessment. Both efficacy-related and safety-related outcomes were considered, providing a more balanced overview of the available evidence. These methodological features support the transparency and reproducibility of the review, while contributing to a clearer understanding of the current evidence and remaining research gaps. Several relevant RCTs are still ongoing due to the emergent nature of the research question.

Conclusion

This systematic review suggests that olanzapine may have orexigenic effects in patients with cancer, with possible improvements in appetite and/or reductions in anorexia even with short-term use, and potential effects on body weight gain with long-term administration. However, the evidence base is limited and should be interpreted cautiously given the heterogeneity across studies, variability in study design and methodological quality. Overall, the findings indicate a promising signal but do not yet support firm therapeutic recommendations. Further well-designed studies are needed to clarify the long-term effects of olanzapine on energy intake, QoL and safety in patients with cancer, and to define the optimal dose and treatment duration for improving appetite and body weight in patients with cancer.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Marie Charlotte Renaux Torres : Conceptualization, Methodology, Formal analysis, Investigation, Data Curation, Writing-Original Draft, Writing-Review & Editing, Visualization, Supervision, Project administration; Kamélia Smati : Investigation, Data Curation, Writing-Review & Editing; Peter Noujaim : Investigation, Data Curation, Writing-Review & Editing; Salomé Martin : Investigation, Data Curation, Writing-Review & Editing; Elia Gigante : Writing-Review & Editing; Mathias Brugel : Writing-Review & Editing; Claire Carlier : Writing-Review & Editing; Zoubir Djerada : Supervision, Writing-Review & Editing; Marine Perrier : Writing-Review & Editing ;Éric Bertin : Supervision, Writing-Review & Editing; Olivier Bouché : Supervision, Writing-Review & Editing; Claire Coutureau : Conceptualization, Methodology, Validation, Formal analysis, Investigation, Writing-Review & Editing, Supervision, Project administration; Florian Slimano : Conceptualization, Methodology, Validation, Formal analysis, Resources, Investigation, Writing-Original Draft, Writing-Review & Editing, Supervision, Project administration.

Funding

Not applicable.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.