Blog · ppj.es
Fatigue with Adjuvant/Neoadjuvant Immunotherapy: +21% vs Placebo, Patients Notice It
Published on 6 August 2026
🇪🇸 Leer este artículo en español
Immunotherapy with immune checkpoint inhibitors (ICIs) in curative-intent settings (adjuvant/neoadjuvant) increases fatigue by 21% versus placebo (RR 1.21; 95% CI 1.08-1.35) [PMID: 42554880] and patients perceive it (SMD 0.12; 95% CI 0.01-0.23 in PROs) [PMID: 42554880]. Compared with chemotherapy, there is no significant difference (RR 0.81; 95% CI 0.36-1.82) [PMID: 42554880]. This is the first meta-analysis to quantify this in a curative setting [DOI: 10.1007/s00520-026-10996-1].
Study design
Systematic review + meta-analysis (PRISMA, Cochrane, GRADE) [PMID: 42554880]. Searches in 4 databases + ClinicalTrials.gov through Sept 2024. 43 studies (19 phase 2, 24 phase 3; 98% randomized), 17,098 patients in the main comparison (ICI monotherapy vs placebo/no treatment) [PMID: 42554880]. Cancers: melanoma (21%), triple-negative breast (16%), NSCLC (16%), renal (9%), gastroesophageal (9%) [DOI: 10.1007/s00520-026-10996-1]. ICIs: pembrolizumab (33%), nivolumab (16%), nivo+ipi (14%), durvalumab (12%), atezolizumab (12%) [PMID: 42554880]. Fatigue measured by CTCAE v5.0 (98% of studies) and PROs —mainly EORTC QLQ-C30— (23%) [PMID: 42554880].
Numerical results (integrated critical appraisal)
| Comparison | Result | Studies (n) | Heterogeneity | GRADE certainty |
|---|---|---|---|---|
| Fatigue any grade (CTCAE) | ||||
| ICI mono vs placebo | RR 1.21 (1.08-1.35) [PMID: 42554880] | 28 (17,098) | I²=78% | Moderate ↓ |
| ICI mono vs chemotherapy | RR 0.81 (0.36-1.82) [PMID: 42554880] | 2 (868) | I²=78% | Very low ↓↓↓ |
| ICI mono vs interferon α | RR 0.63-0.71 (favors ICI) [PMID: 42554880] | 2 (749) | I²=0% | Low ↓↓ |
| PD-1 vs CTLA-4 (fatigue any grade) | RR 1.11 (1.01-1.21) [PMID: 42554880] | 2 (2198) | I²=0% | Moderate ↓ |
| Fatigue grade 3+ | ||||
| ICI mono vs placebo | RR 1.89 (1.32-2.71) [PMID: 42554880] | 18 | I²=65% | Moderate ↓ |
| PROs (fatigue level, SMD) | ||||
| ICI mono vs placebo | SMD 0.12 (0.01-0.23) [PMID: 42554880] | 3 (KEYNOTE-091, CheckMate 238, EORTC 18071) | I²=30% | Moderate ↓ |
| Ipilimumab (SMD) | 0.22 (0.01-0.43) [PMID: 42554880] | 1 (EORTC 18071) | — | Low ↓↓ |
| Global quality of life (SMD) | ||||
| ICI mono vs control | -0.06 (-0.14 to 0.02) [PMID: 42554880] | 6 (4203) | I²=40% | Moderate ↓ |
| Total treatment discontinuation | ||||
| ICI mono vs placebo | RR 1.22 (1.08-1.39) [PMID: 42554880] | 26 (16,423) | I²=89% | Low ↓↓ |
| ICI combo vs ICI mono | RR 1.98 (1.58-2.50) [PMID: 42554880] | 5 (735) | I²=0% | High |
| Discontinuation due to adverse events | ||||
| ICI mono vs placebo | RR 4.84 (3.34-6.99) [PMID: 42554880] | 21 (14,546) | I²=84% | Moderate ↓ |
| PD-1 vs CTLA-4 | RR 0.28 (0.17-0.46) [PMID: 42554880] | 2 (2198) | I²=88% | High |
| ICI combo vs ICI mono (due to AE) | RR 2.44 (1.80-3.30) [PMID: 42554880] | 5 (735) | I²=0% | High |
Explicit critical appraisal
What the paper DOES tell us
- Fatigue is real and measurable in curative-intent treatment: +21% incidence (CTCAE) [PMID: 42554880] and +0.12 SMD (PROs) [PMID: 42554880]. It is not “just quality of life”; it is a discontinuation criterion (RR 4.84 due to AE) [PMID: 42554880].
- PD-1 (pembro/nivo) causes more fatigue than CTLA-4 (ipi) [PMID: 42554880], but ipi AE-related discontinuation is 3.6x higher (RR 0.28 PD-1 vs CTLA-4) [PMID: 42554880]. The toxicity profile differs.
- ICI + chemotherapy ≈ chemotherapy alone for fatigue (RR 0.81, NS) [PMID: 42554880]. Chemo already causes fatigue; adding ICI does not significantly worsen it in the 2 available studies (PEARLS/KEYNOTE-091, KEYNOTE-671) [PMID: 42554880].
- ICI+ICI combination doubles discontinuation (total RR 1.98, due-to-AE RR 2.44) [PMID: 42554880] vs monotherapy. No signal of extra fatigue in combo vs placebo (few studies), but overall toxicity does increase [DOI: 10.1007/s00520-026-10996-1].
- Global quality of life does NOT worsen (SMD -0.06, NS) [PMID: 42554880]. Fatigued patients do not perceive worse global health —a frequent paradox in oncology [PMID: 42554880].
What it does NOT tell us / major limitations
| Gap | Clinical impact |
|---|---|
| Only 3 studies with PROs (of 43) [PMID: 42554880] | SMD 0.12 based on KEYNOTE-091, CheckMate 238, EORTC 18071 —sample biased toward melanoma/lung |
| I²=78% in primary outcome [PMID: 42554880] | High unexplained heterogeneity —different cancers, ICIs, schedules, follow-ups |
| High risk of bias (lack of blinding in almost all) [PMID: 42554880] | Fatigue is subjective; not blinding inflates RR —GRADE certainty downgraded |
| Short follow-up (median ~1-2 years) [PMID: 42554880] | Chronic post-ICI fatigue unknown —may persist for years (as with post-chemo) |
| Only 7 TNBC studies (16%) [PMID: 42554880] | Extrapolation to your population (HR+/HER2- breast with adjuvant ICI) is weak |
| No data on severity/trajectory [PMID: 42554880] | RR 1.21 pools grade 1 + grade 3 —clinically very different |
Publication bias
Symmetric funnel plot (Egger p=0.24) [PMID: 42554880] —but only 28 studies in the main meta-analysis; limited power to detect asymmetry [DOI: 10.1007/s00520-026-10996-1].
Direct implication for your clinic (and for TOXMON)
1. Informed consent — data for the conversation
“With adjuvant/neoadjuvant immunotherapy, 1 in 5 more patients reports fatigue versus not receiving it (RR 1.21) [PMID: 42554880]. Most is grade 1-2; grade 3+ occurs at ~2x versus placebo [PMID: 42554880]. Compared with chemotherapy, fatigue is similar [PMID: 42554880]. Global quality of life does not change [PMID: 42554880].”
2. PRO-CTCAE screening (fatigue item) — mandatory in TOXMON
- Baseline (pre-ICI)
- Every cycle / every 3 weeks
- Alert threshold: worsening ≥2 grades or grade ≥2 persisting >2 weeks
- Referral to psycho-oncology / supervised exercise / CBT (ASCO-SIO 2024 guideline) [PMID: 42554880]
3. Evidence-based interventions (ASCO-SIO 2024 / NCCN 2026) [PMID: 42554880]
| During ICI | Post-ICI / Survivorship |
|---|---|
| Supervised exercise (Grade A) [PMID: 42554880] | Exercise + CBT + mindfulness (Grade A) [PMID: 42554880] |
| CBT / mindfulness (Grade B) [PMID: 42554880] | Yoga, acupressure, moxibustion (Grade B) [PMID: 42554880] |
| American ginseng (Grade B) [PMID: 42554880] | — |
| NO: L-carnitine, antidepressants, routine psychostimulants [PMID: 42554880] |
4. Discontinuation: the number that changes the decision
- PD-1 mono: AE-related discontinuation ~10-15% (vs 2-3% placebo) [PMID: 42554880] —manageable
- CTLA-4 mono (ipi 10 mg/kg): AE-related discontinuation ~35-40% [PMID: 42554880] —reassess indication
- ICI+ICI combo: AE-related discontinuation 2.44x mono [PMID: 42554880] —trials only / strict selection
One-sentence conclusion
Curative-intent ICI adds mild-moderate fatigue (RR 1.21 [PMID: 42554880], SMD 0.12 [PMID: 42554880]) but does not worsen global quality of life [PMID: 42554880]; PD-1 is better tolerated than CTLA-4 [PMID: 42554880]; combo doubles discontinuation [PMID: 42554880]. Systematic PRO screening + early exercise/CBT are standard of care.
Critical appraisal — summary table for the file
| Item | Value | Comment |
|---|---|---|
| PICO question | Adjuvant/neoadjuvant ICI vs placebo/chemo → fatigue | Well defined, clinically relevant [PMID: 42554880] |
| Design | SR + MA (Cochrane, PRISMA, GRADE) [PMID: 42554880] | Robust methodology |
| Population | 43 studies, 17k pts (primary), multiple tumors [PMID: 42554880] | Heterogeneous —variable applicability |
| Primary outcome | Fatigue any grade (CTCAE) + PROs (EORTC QLQ-C30) [PMID: 42554880] | Dual perspective: clinician + patient |
| Primary result | RR 1.21 (1.08-1.35) [PMID: 42554880]; SMD 0.12 (0.01-0.23) [PMID: 42554880] | Significant but small-moderate effect |
| Evidence certainty | Moderate (↓ for blinding bias + high I²) [PMID: 42554880] | Not high —provisional conclusions |
| Applicability to your practice | Moderate —few data on HR+/HER2- breast with adjuvant ICI [PMID: 42554880] | Use as framework, not as direct guideline |
| Major bias | Lack of blinding in subjective fatigue + I²=78% [PMID: 42554880] | Likely overestimation of RR |
Source: Potter L, Lopez-Olivo MA, Uppal R, et al. Cancer-related fatigue during treatment with neoadjuvant and/or adjuvant immune checkpoint inhibitors: a systematic review and meta-analysis. Support Care Cancer. 2026;34:829. DOI: 10.1007/s00520-026-10996-1. PMID: 42554880. Open Access.
Link: https://link.springer.com/article/10.1007/s00520-026-10996-1 (free PDF)
Supplementary data: OSF io/5zfrk (protocol + extracted data)
— This analysis was generated by ANGIE (Always Next to Guide, Inspire and Empower), an artificial intelligence system with SOUL profiles, designed by Dr. Javier Pumares Pérez.
Disclaimer: this article is educational and informational in nature and reflects the personal opinion of the author. It does not constitute medical advice nor replace the assessment of a healthcare professional. If you have a health concern, consult your physician.