Malignant pleural effusion (MPE) is the build-up of fluid between the lung and the chest wall as a result of cancer cells in the pleura. MPE is a common complication of cancer, with an incidence of 50 000 per year in the UK (1) and occurs in up to 15% of people with cancer. MPE can be associated with any type of cancer, both primary pleural malignancy (mesothelioma) and the result of secondary spread from other sites including lung, breast and ovarian (2). The effects upon patients living with MPE are profound, including significant breathlessness, fatigue and impact on daily activity (3). Furthermore, MPE is typically associated with poor prognosis and a median survival of 3-12 months (4). Recent data has indicated that the impact of MPE on healthcare is substantial, with the estimated annual national cost in the USA surpassing $1.5 billion and hospital readmissions leading to costs of $236 million annually (5). Over the past decade, the management of MPE has progressed significantly with an ever increasing number of high quality randomised trials to guide best practice (6–9). Despite the improving evidence base, a number of challenges persist in this vulnerable patient population including optimising time to diagnosis and definitive fluid control. The issue of survival in MPE is of great importance as it informs patient and physician decision making regarding management strategies. The importance of accurate survival scores is amplified in this cohort of patients, in whom balancing the short survival time with volume of hospital contact for fluid management is paramount. Nonetheless, prognostication has been amongst the more challenging aspects of MPE management, due to significant heterogeneity in both underlying malignancy and performance status of patients.

This article reviews the recent advances and standards of care in the management of MPE, while addressing the challenges and key areas requiring further targeted studies over the coming years to optimise the care of patients suffering with MPE.

The current investigation and management pathway for a new pleural effusion is described in Figure 1

The pathway begins with a symptomatic patient presenting to either primary or secondary care with breathlessness, and basic imaging (chest radiograph) demonstrating a unilateral pleural effusion. The priorities for the patient and clinicians are to 1) establish a diagnosis while also 2) providing relief of symptoms. The initial procedure involves aspiration of pleural fluid with around 50mls sent for laboratory diagnostic analysis and assessment of cytology to establish a malignant diagnosis. In addition, a further 1-1.5 litres of fluid may be withdrawn to improve breathlessness.

Recent evidence suggests that the initial pleural aspiration may have limited utility in the diagnosis of MPE (10). The sensitivity of pleural fluid alone is low; even when malignant cells are detected, the sample may be insufficient to provide information to guide oncological treatment (10, 11) (‘actionable histology’), and the fluid recurs in the majority of patients. Following this first procedure, the patient therefore requires further procedures to achieve a diagnosis (pleural biopsy), and a further ‘definitive’ pleural fluid control procedure. This is conducted using either chemical pleurodesis to seal the pleural space or indwelling pleural catheter insertion (IPC) to control breathlessness and prevent re-admission to hospital. The relative merits and risks of these methods are evaluated below.

Recurrent pleural effusion caused by malignancy is often a debilitating condition for patients, with adverse effects on activity levels and performance status (36). The effects of MPE can also impact patients’ tolerance for systemic therapy resulting in a vicious cycle of pleural effusion build-up leading to lack of disease control measures such as missed systemic treatment. As such, patients and physicians are encouraged to openly discuss definitive management options for MPE, which despite great progress in recent years, remains palliative. The historical choice for prevention of pleural fluid recurrence has been chemical pleurodesis, with the greatest evidence base for graded talc as the agent of choice (37, 38). The chemical sclerosant precipitates a diffuse inflammatory reaction and fibrin deposition between parietal and visceral pleura, thus obliterating the pleural space and preventing fluid reaccumulation. The process typically requires an inpatient hospital stay for chest drain insertion and complete fluid drainage and is successful in approximately 70% of cases (6, 35). Although initially the success rate of pleurodesis was previously felt to be greater using talc poudrage at thoracoscopy, the TAPPS trial has shown pleurodesis failure rates at 90 days to be equivalent between poudrage and talc instillation via a chest tube (35).

IPCs are a silicone tube tunnelled under the subcutaneous tissue into the pleural space to allow ongoing drainage of pleural fluid in the patients home, by the patient themselves or a carer/nurse. IPCs carry some major advantages over talc pleurodesis, requiring only a day case procedure under local anaesthetic and providing equally effective management of MPE in patients with NEL. Over the past decade, IPCs have become the subject of a number of rigorous, high quality randomised clinical trials, with in depth evaluation of effectiveness in symptom control and cost effectiveness. Over this period these trials have not only seen a paradigm shift in the options offered to patients to manage MPE, but also the use of patient facing outcome measures (PROMS) as the primary outcome in the aforementioned clinical trials. These trials have shown that IPCs improve breathlessness and quality of life comparably to talc pleurodesis and reduce the length of hospitalisation by 2 days (6, 7). On this evidence alone, IPCs have become a viable option to patients with recurrent MPE, allowing a choice between IPC and talc according to patient preference. Whether IPCs should become a first line intervention over talc pleurodesis is however a more nuanced issue. IPC related complications occur in approximately 10-20% of patients, although the majority are minor and treatable, such as cellulitis, with true infection of the pleural space reported to be less than 5% (39, 40). In addition to this, the genuine impact of long term IPC drainage on both the patient and healthcare systems is significantly under-studied, as hospital admission days do not account for the limitations of requiring domiciliary drainage, often by a trained nurse. Even cost effectiveness studies do not clearly favour one intervention, with a post hoc analysis of the TIME-2 trial revealing that for patients with survival limited to <14 weeks, IPCs were more cost effective, however beyond this, talc pleurodesis was favourable (41).

Thus the choice between talc pleurodesis and IPC remains dependent on patient choice, resource availability and physician familiarity at the current juncture, although with the advent of combined treatments (see below), the balance of these factors may yet change. Spontaneous pleurodesis (autopleurodesis) has been reported with IPCs, and retrospective datasets have suggested the rate of autopleurodesis with IPCs to be 43-47% (37, 42), although in prospective studies suggest this is somewhat lower (24-27%) (9).

Figure 2 illustrates a suggested evidence based flowchart for the management of MPE, and the question future studies will seek to target is whether waiting times to definitive fluid control can be minimised or even delivered as part of a ‘first intervention’ for suspected MPE.

To date, prognostication in MPE has been addressed by two prognostic scoring systems validated in MPE; the LENT and PROMISE scores. The LENT score was derived using 3 prospectively collected datasets and retrospectively derived. LENT assigns patients to a low, moderate or high risk of death (319, 130, 44 days median survival respectively) using pleural fluid LDH, ECOG score, blood neutrophil-to-lymphocyte ratio (NLR), and tumour type (47). In a validation cohort, the LENT score was found to perform significantly better than ECOG performance status alone in predicting survival.

The PROMISE study stratified patients into four survival categories at three months ranging from <25% to >75%. The score includes clinical parameters (ECOG performance score, previous chemotherapy and radiotherapy, cancer type) and biological variables (white blood cell count, C-reactive protein, haemoglobin and serum levels of tissue inhibitor of metalloproteinases-1, TIMP-1) (48). Despite their simplicity (with the exception of TIMP-1 in PROMISE which is not routinely measured in clinical practice) and external validation, there has been suggestion that alternative scoring systems are necessary to correct for regional demographics variation, for example in areas with high rates of EGFR adenocarcinoma mutations (49) and target specific tumour types. As such one recent study has sought to address this using disease specific models to allow more precision in survival prediction. The breast and lung effusion survival score (BLESS) was derived retrospectively from analysis of 562 patients, and validated in a separate cohort of 727 patients. Both the lung and breast models utilise variables of ECOG performance status, benign pleural fluid cytology, pleural fluid LDH and pleural fluid protein. The lung model adds history of surgery within 30 days and the presence of bilateral pleural effusions. The breast model adds NLR. The authors concluded that in lung and breast malignancy, the BLESS score outperformed LENT, adding another potential tool to the prognostication of MPE (50). It remains to be seen whether these scoring systems become widely utilised in clinical practice however, as no studies to date have demonstrated clinical impact on patient reported outcomes or the clinical pathway in MPE.

An area with great potential to progress the management of patients with MPE is that of more sophisticated biomarker prediction of clinical outcomes including fluid volume prediction and autopleurodesis. Other fields have successfully integrated biomarker driven care pathways (51) and this remains lacking in MPE and pleural medicine in general. Important data has been discovered as part of the PROMISE study which aimed to discover and validate pleural fluid biomarkers to predict outcomes. Despite the analysis of over 1200 proteins, only 4 showed significant association with survival – TIMP-1, VCAN, GSN and MIF. Of note however, none of these could predict pleurodesis success, which for patients may represent a more direct impact on choice of fluid management strategy.

In regards to predictors of fluid output and autopleurodesis, early data suggests that routine clinical laboratory tests are not helpful in predicting outcome (42). A recent study has identified Vascular Endothelial Growth Factor (VEGF), Transforming Growth Factor-B (TGF-B) and Basic Fibroblast Growth Factor (FGF2) as key players in auto-pleurodesis induced by IPC (52). However, this was a longitudinal study and further studies are needed to assess these findings. A blind exploratory study to screen for auto-pleurodesis regulators has not been performed, which is a necessary next step to objectively identify the underlying molecular mechanisms underpinning autopleurodesis.

Over the last decade, great progress has been made in the treatment of MPE, with a formerly reactive and opportunistic approach moving to a robust evidence-based paradigm for both diagnostic and therapeutic options. Despite this, there remains key gaps in the evidence base. The true patient experience in MPE requires further assessment and the delivery of high yield diagnostics and definitive fluid control should be evaluated to determine if patients can progress to systemic treatment and symptom control earlier than current international guidelines allow. Over the next decade, moving beyond drainage strategies to biomarker and immunological analysis of MPE formation and recurrence will be essential, and may even lead to targeted pharmacological treatment of malignant pleural effusion.

DA was responsible for initial draft preparation and revision. All other authors were responsible for reviewing and approving the final manuscript. All authors are responsible for the overall content.

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.

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.