Lung cancer is the second most common cancer worldwide and accounts for 25% of all cancer deaths (1). Non-small cell lung cancer (NSCLC) represents approximately 85% of all cases. Unfortunately, the majority present with metastatic disease for whom the prognosis is often poor (2). Historically, NSCLC was considered as a single disease, and patients with metastatic disease were treated with cytotoxic chemotherapy. Advances in our understanding of the biology of NSCLC have defined molecular subtypes of NSCLC with distinct clinical characteristics and therapeutic options. Small molecule tyrosine kinase inhibitor (TKI) “targeted therapies” are available for the 10–15% of patients with NSCLC harbouring epidermal growth factor receptor (EGFR) mutations, i.e., erlotinib, gefitinib, afatinib, and osimertinib, and the 5–6% harbouring anaplastic lymphoma kinase (ALK) gene rearrangements or c-ros oncogene 1 (ROS1) mutations, i.e., crizotinib and alectinib (3–5). In patients without such actionable mutations, the monoclonal antibody “immunotherapy” agents which inhibit immune checkpoints, i.e., pembrolizumab, nivolumab, atezolizumab, and durvalumab, have emerged as standard-of-care treatment (6–12).

Clinical assessment of the fitness of a patient, using tools such as the Eastern Co-Operative Group Performance Status (PS), and consideration of the comorbidities of a patient determine whether it is appropriate to treat an individual with these agents. The identification of objective prognostic factors that predict response to systemic therapy in NSCLC would be invaluable tools for clinical decision-making. In particular, factors that identify those that are least likely to respond to systemic therapy or that predict the development of treatment resistance may allow patients to pursue other treatment options or facilitate early referral to palliative care and appropriate arrangements for end of life care.

Inflammation plays a key role in the development, survival, and progression of cancer (13). In addition to interactions at the level of the tumour microenvironment, systemic inflammatory effects are observed and frequently measured by routine clinical tests such as serum albumin. Historically, albumin and other serum proteins have been widely used by physicians as biomarkers of malnutrition (14–16). However, serum levels of this cheap, readily available blood biomarker are significantly depressed as a result of inflammation through decreased synthesis and increased catabolism (16, 17). As such, albumin is considered a biomarker of cachexia, a state characterised by weight loss, systemic inflammation and reduced functional status and associated with poor prognosis in many diseases (18–20). Indeed, hypoalbuminaemia is a recognised poor prognostic factor in cancer, both individually and as a core component of composite prognostic scores (15, 21–24). However, much of the established literature for the use of albumin as a biomarker in NSCLC pre-dates the routine use of immunotherapy and molecular profile directed targeted therapy treatment in clinical practise (24–27).

The Edinburgh Cancer Centres provide regional cancer services in Scotland, UK, serving a population of approximately 1.5 million. We utilised real-world experience to examine the prognostic value of serum albumin concentration (g/L) in patients with metastatic NSCLC receiving first-line targeted therapy or immunotherapy-based systemic anticancer therapy (SACT).

Biomarkers of systemic inflammation and malnutrition, such as albumin, are firmly established as having a prognostic value in patients with cancer. Hypoalbuminaemia is an important prognostic factor for patients with NSCLC, including in patients with localised disease treated with surgery or those with metastatic disease treated with cytotoxic chemotherapy (22–25, 27–30). This study found that albumin independently predicts survival in patients with metastatic NSCLC treated with either first-line targeted therapy or immunotherapy-based treatments. This is an important finding given the distinct molecular and clinical characteristics of these subtypes of NSCLC. In particular, first-line immunotherapy-based treatment regimens show limited efficacy in patients with actionable mutations (11, 31–33).

The key strength of our study is in the simplicity of utilising a single routine blood test that provided useful prognostic information in well-defined clinically relevant patient groups based on “standard of care” molecularly defined treatment pathways. Although albumin has previously been shown to predict survival in patients with metastatic NSCLC treated with the EGFR targeted therapy erlotinib, only 6.8% of patients had a known EGFR mutation and 90% had previous exposure to other SACT agents (27). Similarly, although several studies had investigated the prognostic value of albumin in patients with NSCLC treated with immunotherapy, many have combined albumin with other factors, and none have restricted assessments to first-line treatment only (34–37).

Despite the suggestion made by the standard pre-treatment assessment that all patients in this study were fit for SACT, more than 10% of patients died within 12 weeks of starting treatment. This timescale is clinically relevant as patients typically undergo first formal assessment of treatment efficacy at this time and expected life expectancy of at least three months was a key eligibility criterion in the clinical trials of these agents (6–8, 38). In both the targeted therapy and immunotherapy cohorts, albumin <35 g/L was associated with a significantly higher risk of death within 12 weeks of starting treatment. A key aim of pre-treatment prognostic biomarkers is to identify patients who will not benefit from treatment to facilitate personalised treatment options and enable realistic discussions with patients about expected outcomes. Specifically, this may reduce the risks associated with treatment and enable appropriate palliative care and end-of-life decision making in a planned manner.

Significantly, few studies had investigated the prognostic value of biomarkers of systemic inflammation beyond pre-treatment assessment. The neutrophil/lymphocyte ratio (NLR) or platelet/lymphocyte ratio (PLR) during or after treatment have been shown to be associated with survival in patients with NSCLC treated with SACT (24, 39–42). An increase in NLR ≥3 from pre-treatment to 6 weeks is associated with poorer outcomes in patients with NSCLC treated with immunotherapy (42). In our study, although dynamic changes of albumin status were associated with differences in survival, the prognostic value of 12-week albumin was independent of pre-treatment albumin. In patients who were alive at 12 weeks, albumin <35 g/L was associated with a worse ongoing prognosis in both cohorts.

These findings also suggested that albumin has ongoing prognostic value during the treatment journey of a patient. An improvement in albumin levels while on treatment may represent better cancer control. Whether this is by reducing cancer activity directly, or indirectly by reducing inflammation is not clear. By extension, declining albumin may represent a loss of cancer control. This may have novel clinically relevant implications. Firstly, serial measurement of albumin, and potentially other biomarkers of systemic inflammation, may be useful as tools for monitoring treatment response. Declining albumin status while on treatment may help identify patients who are developing treatment resistance. In clinical practise, this may support decisions for earlier formal response assessments or timelier switches to alternative therapy. This is strongly supported by our finding that a majority of patients in each cohort had hypoalbuminaemia at the time of progressive disease. Progressive disease-albumin <35 g/L was associated with a median survival of <6 weeks in each cohort. Therefore, these patients were unlikely to derive benefit from second-line therapy, which should not be recommended. To our knowledge, we are the first to assess the prognostic value of biomarkers of systemic inflammation at the time of disease progression in these patient cohorts.

Secondly, these findings strengthened the growing evidence promoting the use of rehabilitation in patients with cancer. Albumin, like other biomarkers of systemic inflammation, reflects the multi-factorial syndrome of cachexia (19, 20). Strategies to stabilise patient fitness and reduce systemic inflammation/cachexia may have important roles alongside SACT in the holistic management of patients with cancer (18, 43). Recently published guidelines in the UK have recommended integrating pre-treatment rehabilitation, i.e., prehabilitation, into established clinical pathways for patients with cancer (44). Interventions to optimise physical functional and nutritional status may be employed based on the needs of the patient (18, 43–46). Patients with metastatic NSCLC often have multiple pre-existing comorbidities and a high cancer-symptom burden (47). In our cohort, 64 (16%) had pre-treatment PS2+, which is consistent with that seen in previous real-world studies (48). Overall, this is a patient group who were likely to require intensive, specialist, multi-modal intervention to maximise their fitness. For example, recent data has suggested that half of the patients with metastatic NSCLC who are fit for SACT meet the criteria for the critical need to see a dietician (47).

Interventional rehabilitation studies for patients with cancer are in progress (45, 49). The use of systemic therapies to improve muscle mass and reduce cachexia are also being trialled, although none have entered routine clinical practise (50, 51). We strongly advocate the evaluation of the impact of rehabilitation interventions on biomarkers of systemic inflammation and the relationship to outcomes in patients with metastatic cancer receiving SACT. In the meantime, we suggested that these biomarkers may have roles to play in selecting patient groups for rehabilitation, stratifying interventions and monitoring response.

The results of the present study confirmed that albumin is an important prognostic factor in patients with metastatic NSCLC, predicting survival independent of the class of drug treatment based on molecular profiling. Most importantly, we provided new evidence to show that tracking albumin during therapy may reveal information about disease activity and treatment response over time. Further work is needed to confirm whether strategies aimed at improving albumin will lead to improved outcomes in patients with metastatic NSCLC receiving SACT.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

MS and IP conceived the idea. MS, AS, KC, T-ED, JL, FT, and CS led the primary data collection. MS, AS, KC, and T-ED were responsible for data cleaning. MS analysed the data. MS, AS, KC, and IP prepared the draft manuscript. T-ED and MM provided significant intellectual input and advice in the re-draft of the manuscript. All authors approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The handling editor declared a shared affiliation with several of the authors (MS, CS, FT) at time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.