Biomarkers in Onco-Critical Care: Understanding the Role of C-reactive Protein and Procalcitonin in Infection versus Tumour-induced Inflammation

Acute phase reactants have been utilised extensively in critical care units for direct and surrogate diagnosis of systemic infection, febrile neutropenia and systemic inflammatory response syndrome. But these reactants also can increase in the face of tumour lysis, blood transfusion reaction and as an interaction with selective chemotherapeutic drugs. To confuse the above scenarios, malignancy by itself can increase acute phase reactants. Solid organ transplant surgeries add further complexity to the nature of the rise in acute phase reactant proteins, indicating infection versus graft rejection. Bone marrow transplant recipients further add up with underlying neutropenia, associated with or without fever and sepsis. Hence, the utility of these biomarkers in perioperative onco-anaesthesia becomes even more perplexing, owing to infectious and non-infectious aetiologies. C-reactive protein (CRP) and procalcitonin (PCT) have been extensively investigated in leukaemia, lymphoma and solid organ tumours.[1] PCT a pro-hormone of calcitonin is secreted under normal conditions by the thyroid medulla, but during infection gets secreted by extra-medullary sites including liver, kidney, lung and inflamed tissues similar to cytokine release, leading to the terminology of ‘hormokine’ with a half-life extending over 25–30 h [Figure 1].[2,3]Figure 1: (a) Synthesis, secretion and expression of procalcitonin in the body, (b) Elevation of procalcitonin in different states. PCT: ProcalcitoninMalignancy superadded with inflammation-infection and elevated CRP were evaluated to obtain a causal demonstration of raised PCT in an observational study.[1] Fever of unknown origin, high tumour load and drug interactions contributed to the conflict in an otherwise demonstrable infection upon cultures on blood or body fluid. Radiological imaging in the form of chest radiography, ultrasound abdomen and computed tomography were evaluated besides obtaining cultures.[1] The median PCT and CRP were evaluated in the infection group, non-infection group and unexplained fever groups. Cytosine arabinoside and all-trans retinoic acid were commonly associated with fever in the drug reaction group (along with transfusion risk); whereas bulky solid organ tumours as well as non-Hodgkin lymphoma were inclusions for tumour load group. PCT demonstrated significance in the infection group when compared against the non-infection group, drug reaction and tumour load groups; whereas CRP demonstrated significance only among infection and drug reaction groups.[1] PCT elevation was demonstrable in both tumour load as well as tumour lysis. The eventual cytokine release, especially after therapy with monoclonal antibodies (immunotherapy) has been considered to conflict with the interpretation of infection.[1] Further leucocyte count also determines the extent of PCT rise in concentration; elevated counts coinciding with elevated PCT (normal PCT cutoff 0.5 µg/L or 0.5 ng/mL) and leucopenia demonstrating lower PCT concentrations.[1,4] CRP being a sensitive marker for inflammation is restricted by drug-induced elevated concentrations leading to its limited specificity. High CRP levels (normal CRP cutoff 85.1 mg/L) are associated with infection as well as in tumour load, thereby conflicting with its interpretation in haemato-oncological diseases associated with a high tumour load burden. CRP concentrations could not differentiate between infective and patients without infection thereby hindering clinical decision-making.[1] A combination of PCT and CRP proved beneficial in the subset of patients without infection, wherein PCT in an already elevated CRP among patients without infection was found to be low (≤0.5 µg/L). In patients presenting with para-neoplastic febrile response, the absence of infection could be established with low PCT levels despite an increased CRP.[5] The combination of CRP and PCT in clinical situations arising due to drug-related or tumour-related febrile response could be discriminated by PCT elevation only in the presence of infection.[6] The interpretation of PCT should be done with caution in patients already presenting with co-existing leucopenia/neutropenia as PCT may be low despite an infection.[7] Similarly, PCT levels are elevated in small cell lung cancer despite the absence of any infection, for which it could not live up as a prognostic marker in this subset.[8] Similarly, PCT concentrations were elevated in PCT-secreting hepatocellular carcinoma, which became normalised after resection of the involved segment.[9] Medullary carcinoma of thyroid and neuroendocrine tumours also present with elevated PCT concentrations, mimicking an infection.[2] The baseline PCT levels may often be increased in case of advanced cancer and metastasis, contributing to the necessity to obtain serial PCT values, being considered to be a predictor associated with advanced malignancy and metastatic disease states without underlying clinical infection, for which interleukin 6 and fluorodeoxyglucose positron emission tomography would further complement.[10] PCT was found elevated in patients with underlying infection and metastasis group rather than in patients with localised infection and early cancer.[11] PCT has been shown to reliably increase in febrile neutropenia with underlying sepsis, rather than in febrile neutropenia patients without demonstrable infection (fever of unknown origin).[4,12] PCT has been well-established in detecting infection in paediatric and young adult patients presenting with febrile neutropenia.[13–17] The real test of significance arises in non-neutropenic patients presenting with fever, which has been addressed by a recent meta-analysis, in which it revealed diagnostic accuracy with moderate specificity (78%) but a relatively poor pooled sensitivity (60%), in detecting infection among non-neutropenic febrile cancer subset, indicating that sole reliance on PCT as a biomarker not to be recommended in this cohort.[13] To conclude, the message has been loud and clear, that serial measurements and not single measurement of PCT are predictive of bacteraemia after initiating chemotherapy. Similarly, a decreasing trend in PCT from the baseline, which had predicted bloodstream infection and sepsis in non-neutropenic febrile cancer patients in response to appropriate antibiotic therapy, confirms underlying bacterial infection from that of a tumour-related fever. At the same time, biomarkers alone will not be a replacement for clinical situational judgement, but would invariably contribute by complementing it.