The effectiveness of intensive care unit (ICU) care for cancer patients remains controversial. Advances in antitumour and supportive care led to major improvements in outcomes of oncological patients in the ICU. Improved cancer therapies and supportive management of organ dysfunctions have contributed to improved survival rates. As a consequence, the number of cancer patients requiring ICU admission is rising. Frequently, cancer patients have a poor performance status and are vulnerable. It is a heterogeneous population, where the nature and curability of the neoplasm and the severity of critical illness cause a plethora of issues about ICU admissions. Therefore, oncological patients are often considered inappropriate for ICU admission. So far, no specific severity-of-illness scoring system can reliably predict the outcome of critically ill oncological patients and scoring systems or survival predictors are lacking. The major determinants of mortality and prognosis are the number of organ failures, need of mechanical ventilation (especially for acute respiratory distress syndrome), vasopressor support (>4 hours) and therapies that have preceded ICU admission. The underlying neoplasm seems to have a little impact on the outcome. The most frequent reasons leading a cancer patient to ICU are postoperative recovery, respiratory failure, infection and sepsis. To date, scientific reports suggest that acute organ dysfunction should be managed at its onset, preferably within 2 hours after the admission, whereas further aggressive ICU management should be reappraised after a few days of full support. Prognosis should be reassessed at frequent intervals with particular attention to the development of multiple organ dysfunctions. Discussing the code status is a sensitive matter and should be balanced between the patient's wish and objective medical outcome assessment. The latter can only be achieved in a multidisciplinary team of intensivists, oncologists/haematologists and potentially palliative care experts, preferably by consensus. Transition from restorative to palliative care should be made when there is no benefit from further intensive treatment, there is no trend to recovery in the first days of intensive care and where symptom palliation would improve the quality of life. Patients' autonomy and dignity should remain paramount in any decision-making. Current data do not support any absolute criteria for triaging. Establishment of clear goals and approach to admit and treatment for oncological patients in the ICU are however urgently needed. This requires further prospective studies for independent validation in different medical settings and identifying prognostic tools that can aid with decision-making and patient selection for ICU. Cancer should not be seen as an exclusion criterion and priority should be given to assure the quality of life of oncological patients.