Most Cancer Patients Resume Treatment After ICU, But Half Abandon Original Plan

Surviving a stay in the intensive care unit has long been considered the finish line for critically ill patients with cancer, but a new systematic review suggests it is better understood as a fork in the road. An international team of researchers has compiled the most comprehensive picture to date of what happens to anticancer therapy after an unplanned ICU admission, and their findings complicate the traditional narrative that critical illness marks the end of meaningful cancer care. According to the analysis, published in Supportive Care in Cancer, roughly six in ten patients who survive an ICU stay go on to receive some form of anticancer treatment again. Yet the picture darkens when researchers looked more closely: fewer than half of these survivors actually return to the treatment regimen their oncologists had originally planned before the critical illness struck.

The study, led by Raisa do Val Roso and Antonio Paulo Nassar Junior of the AC Camargo Cancer Center in São Paulo, together with Laveena Munshi of the University of Toronto, addressed a glaring blind spot in oncology and critical care medicine. Patients with cancer now make up approximately one quarter of all ICU admissions, a proportion that has grown steadily as cancer therapies extend survival and create new populations of patients vulnerable to acute complications. Up to 5 percent of patients with solid tumors, and more than 20 percent of those with certain hematologic malignancies, require ICU care within the first year after diagnosis. Hospital survival for these patients has improved dramatically, exceeding 70 percent in many settings, but the question of what happens after discharge has received far less scientific attention than the question of who survives.

To fill that gap, the researchers conducted a systematic review following PRISMA guidelines, with the protocol prospectively registered in PROSPERO. They searched PubMed, Embase, Scopus, and Web of Science from database inception through February 24, 2025, applying no restrictions on language, country, or publication date. The search retrieved 2,597 records, of which 51 underwent full-text review and 20 studies ultimately met the inclusion criteria. Eligible studies had to report on the continuity of systemic anticancer treatment, encompassing chemotherapy, radiotherapy, immunotherapy, targeted therapy, hormone therapy, surgery, or hematopoietic stem cell transplantation, after an unplanned ICU admission in adults with solid or hematological malignancies. Studies limited to pediatric populations or to patients who had decisions made to forgo life-sustaining therapies were excluded.

The 20 studies comprised 5,948 patients in total, of whom 3,551 were specifically assessed for treatment resumption after discharge. Fifteen studies focused exclusively on solid tumors, three on hematological malignancies, and two included both categories. The overwhelming majority were single-center investigations, twelve of the twenty were conducted in France, and only one was prospective in design. The main reasons for ICU admission were infection or sepsis and acute respiratory failure, and the proportion of patients requiring invasive mechanical ventilation ranged widely, from 8.8 percent to 67.7 percent across studies. Notably, only a minority of patients had poor performance status before admission, suggesting that many were candidates for continued cancer therapy in principle.

The headline result of the meta-analysis was a pooled treatment resumption rate of 61 percent, with a 95 percent confidence interval of 51 to 70 percent, within six months of ICU or hospital discharge. The researchers used a random-effects model to pool proportions, an appropriate choice given the anticipated clinical and methodological diversity among the studies, and quantified heterogeneity with the I-squared statistic, which reached a striking 95.6 percent. That figure signals that the individual studies differed enormously from one another, a point the authors acknowledge candidly. Publication bias was assessed through funnel plots and Egger’s linear regression test and found to be absent, with a p-value of 0.91, lending some statistical reassurance to the pooled estimate despite the heterogeneity.

When the analysis was stratified by malignancy type, an intriguing pattern emerged. Patients with solid tumors resumed some form of anticancer treatment at a pooled rate of 60 percent, based on data from 2,772 patients, while patients with hematological malignancies resumed treatment at a notably higher rate of 77 percent, based on 423 patients. The authors offer several possible explanations for this difference. Critical illness in patients with blood cancers often arises directly from the malignancy itself and may respond to urgent treatments such as emergency chemotherapy, which is sometimes even delivered in the ICU. There may also be cultural and specialty-specific differences in how hematologic oncologists and solid tumor oncologists approach post-ICU treatment decisions, a divergence documented in prior surveys of physician attitudes. Given the substantial heterogeneity, however, the authors caution that this difference requires confirmation in larger prospective cohorts.

The secondary outcome told a more sobering story. Eight studies encompassing 1,736 patients specifically assessed whether survivors returned to the exact treatment regimen planned before ICU admission. The pooled rate was just 48 percent, with a confidence interval of 36 to 60 percent, at six months post-discharge. For patients with solid tumors, the rate of resuming the original plan was 46 percent, and for those with hematological malignancies it was 41 percent in the dedicated subgroup analyses. This distinction between resuming any therapy and resuming the planned therapy is clinically meaningful, because modifications to a treatment regimen after critical illness may reflect disease progression, organ dysfunction, diminished functional reserve, or a recalibrated balance of risks and benefits, all of which carry different prognostic implications.

The review also identified factors that predict whether treatment resumes. Poorer performance status, whether at admission or worsened at discharge relative to baseline, was associated with lower odds of resuming treatment in three studies. Older age was linked to discontinuation in two studies. Additional risk factors for stopping or modifying therapy included delirium during the ICU stay, hyperbilirubinemia, failure of more than one organ, therapeutic limitations, cancer progression already evident at ICU admission, corticosteroid use at admission, and tumor sites in the gastrointestinal tract or lung. These variables sketch a coherent physiological picture: patients whose critical illness leaves them frail, confused, or with failing organs, and whose cancer was already advancing, are the least likely to return to active treatment.

The authors are careful about the limits of what their synthesis can show. Nearly all included studies were single-center and retrospective, treatment continuation decisions are shaped by local practice patterns, and definitions of treatment resumption varied from assessment at ICU discharge to six months afterward, introducing conceptual heterogeneity that a meta-analysis cannot fully resolve. Crucially, no study compared ICU survivors with hospitalized cancer patients who did not require intensive care, so it remains impossible to determine whether the treatment disruptions are a consequence of ICU admission specifically or of hospitalization more broadly. Still, at least one prior study using UK Biobank data found that cancer progression was higher among patients who experienced an ICU admission during hospitalization than among those who did not, hinting that reduced treatment resumption may be one contributing mechanism.

For clinicians, the message cuts in two directions. For intensivists, the data argue that ICU admission does not preclude subsequent anticancer therapy in a substantial proportion of survivors, supporting the value of critical care for patients with curable or treatment-responsive malignancies and countering therapeutic nihilism. For oncologists, the finding that more than half of survivors may not resume their originally planned treatment underscores the need for early reassessment after discharge, covering disease status, functional capacity, and patient preferences. The authors frame ICU admission as a potential turning point in the oncologic trajectory rather than an endpoint, and they call for prospective, multicenter studies with standardized definitions of treatment resumption, predefined assessment time points, and integration of functional and patient-reported outcomes. Until such studies arrive, this meta-analysis offers the clearest quantitative foundation yet for conversations between critically ill patients with cancer, their families, and the physicians deciding what comes next.

Subject of Research: Resumption of anticancer treatment after intensive care unit admission in adults with cancer

Article Title: Impact of intensive care unit admission on the resumption of treatment in patients with cancer: a systematic review and meta-analysis

Article References: do Val Roso, R., Daltro-Oliveira, R., Munshi, L., & Nassar Junior, A. P. (2026). Impact of intensive care unit admission on the resumption of treatment in patients with cancer: a systematic review and meta-analysis. Supportive Care in Cancer, 34(9), Article 915. https://doi.org/10.1007/s00520-026-11135-6

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11135-6

Keywords: intensive care unit, cancer, critical care, systematic review, meta-analysis, treatment resumption, solid tumors, hematological malignancies, oncology, patient outcomes, chemotherapy, post-ICU care