Extended pleurectomy decortication for malignant pleural mesothelioma in the elderly: the need for an inclusive yet selective approach†

Abstract

OBJECTIVES

The median age at diagnosis of patients with pleural mesothelioma in the UK is 73 years. Recent series have shown the feasibility of extended pleurectomy decortication in the elderly, but with continuing debate about the efficacy of this treatment, we reviewed our experience to identify more detailed selection criteria.

METHODS

We reviewed prospectively collected data on all patients from 1999 to 2016 undergoing extended pleurectomy decortication. We compared clinical and pathological outcomes and survival data from patients 70 years and older (≥70 years) with those younger than 70 years (<70 years).

RESULTS

Eighty-two of the 300 (27.3%) patients were ≥70 years of age at the time of surgery. More patients in the elderly group required intensive care postoperatively (6.2 vs 16.7%, P = 0.01) and developed atrial fibrillation (14.4 vs 24.4%, P = 0.05). There was no intergroup difference in length of hospital stay or in in-hospital, 30-day or 90-day mortality. Elderly patients were less likely to receive neoadjuvant (<70 years 21.2%, ≥70 years 11.0%; P = 0.045) or adjuvant chemotherapy (<70 years 45.4%, ≥70 years 29.3%; P = 0.04). Median overall survival was similar: <70 years 14.0 months, ≥70 years 10.3 months; P = 0.29. However, in node-positive patients, survival was poorer in the elderly (13.0 vs 9.1 months, P = 0.05), particularly in those with non-epithelioid tumours (3.8 vs 6.7 months, P = 0.04). On multivariable analysis, age was not a significant prognostic factor, although lack of adjuvant therapy (P = 0.001) and admission to the intensive care unit (P < 0.001) remained poor prognostic factors.

CONCLUSIONS

Although age in isolation should not be an exclusion criterion for extended pleurectomy decortication for mesothelioma, in the elderly, a more rigorous preoperative evaluation of nodal disease and an additional assessment of fitness for adjuvant chemotherapy are recommended.

INTRODUCTION

Survival of patients with malignant pleural mesothelioma (MPM) remains poor, with only one proven line of therapy available: chemotherapy in the form of a platinum and pemetrexed doublet [1]. Despite the ban on the importation of asbestos in the developed world, the incidence of MPM continues to increase worldwide [2, 3]. The use of radical surgery with either extrapleural pneumonectomy (EPP) or extended pleurectomy decortication (EPD) remains a topic of debate. However, several series have shown that long-term survival can be achieved, with varying morbidity and mortality rates, albeit in highly selected patients [4–8].

The current median age at diagnosis of patients with MPM in the UK is 73 years, and the majority of patients are older than 70 years at diagnosis [9, 10]. As an increasing number of older patients, with more comorbidities and reduced physiological reserve, are being referred for consideration of radical surgery, we have moved from performing EPP to EPD in our centre as the operation of choice for macroscopic complete resection [11].

Recently, data have been published regarding the use of EPD in patients older than 70 years of age [12]. This experienced group found that it is possible to safely perform an EPD in elderly patients, with no increase in morbidity or mortality, and with maintained long-term survival outcomes. With the continuing debate regarding the efficacy of EPD, we aimed to determine whether the increasing age of patients undergoing this procedure has influenced outcomes and to determine whether a more detailed selection criteria was appropriate in certain cases.

MATERIALS AND METHODS

Patients

All patients undergoing EPD for MPM from 1999 to 2016 were identified from our institutional database. They were then separated into 2 groups by age at operation: either younger than 70 years (<70 years) or 70 years and older (≥70 years). Because of the increasing age of patients, and differing comorbidity profile of patients referred to our centre with mesothelioma for consideration of radical surgery, an intentional transition from routinely performing EPP to EPD occurred over this time period [11]. During the transition period, the main indications for EPD over EPP were patient age, pulmonary function precluding a pneumonectomy and mediastinal lymph node positivity. One hundred and thirty-four patients underwent EPP during this same time period.

Preoperative patient selection

As a tertiary referral centre, our institution receives referrals from over 20 oncology units throughout the UK. Most patients are referred with a biopsy or cytology-confirmed diagnosis of mesothelioma. If the diagnosis is only suspected, or the histological subtype is not known, patients will undergo a video-assisted thoracoscopic biopsy and talc pleurodesis under general anaesthetic at our centre. This provides a confirmation of diagnosis and histological subtype where required and also allows the surgeon to make an additional direct assessment of resectability intraoperatively. We have changed our policy regarding resection in cases of sarcomatoid disease over the last 17 years and only now offer radical surgery to those with biphasic or epithelioid disease [13].

All patients underwent thoracic computed tomography to determine clinical stage and resectability. Positron emission tomography (PET) scanning is not routinely performed preoperatively, neither is invasive mediastinal staging. The majority of patients performed standard pulmonary function tests, including forced expiratory volume in 1 s and forced vital capacity, unless they had a pleural effusion at the time of referral. However, these test results are often difficult to interpret in the presence of an advanced, restrictive tumour. Renal dysfunction was not a specific exclusion criterion for radical surgery.

Operative technique

All patients underwent an EPD, including by definition, resection of the visceral and parietal pleura, and the pericardium and hemi-diaphragm, unless the tumour was easily resected with macroscopic clear margins from either or both of these structures. We have previously found that the diaphragm is not involved by tumour in up to 30% of operated patients with MPM, and as such, we will preserve the diaphragm in cases where we can ensure an R1 resection [14]. Patients who underwent an R2 resection were not included in this study; macroscopic complete resection (R1) was achieved in all 300 patients. Following EPD, patients were routinely managed in a dedicated thoracic surgical high-dependency unit. Very rarely were there preoperative plans for intensive care unit (ICU) admission postoperatively.

Data collection

Our prospectively updated institutional database and patient case notes were reviewed, and demographic, clinical and pathological data were collected. Clinical disease progression was determined from outpatient records, radiological review or correspondence from referring clinicians. Progression-free survival was calculated from the date of surgery to the first clinical or radiological finding of progression, with patients being censored at the time of their last review or at death. Overall survival was calculated from time of operation to death or to the date of censoring. Mortality data were updated monthly through data collection from local hospital databases, hospital episode statistics (HES) data or direct contact with referring hospitals in cases where patients were not from our local catchment area; the last date of censor was 23 April 2016. For those lost to follow-up, the date of censor was their last follow-up date or known date of death.

Indication for neoadjuvant or adjuvant chemotherapy

The provision of neoadjuvant or adjuvant chemotherapy is at the discretion of the referring oncologist. Given that we receive surgical referrals from over 20 centres around the UK, oncological practices, and therefore the neoadjuvant or adjuvant therapeutic strategies, differ markedly from centre to centre and often patient to patient within centres. Increasing numbers of patients are now receiving 2 cycles of platinum/pemetrexed chemotherapy prior to surgical resection as part of the MARS2 study (NCT02040272). Before this, only a very small number of patients (17.3%) were given chemotherapy treatment prior to resection. In some cases, patients were treated with chemotherapy as the first-line treatment and referred for surgical resection on progression as opposed to receiving true neoadjuvant chemotherapy. Many oncologists choose to reserve adjuvant chemotherapy treatment until there is evidence of disease progression, whereas others will give adjuvant platinum/pemetrexed to all patients or specifically to those with biphasic disease or pathological nodal disease [15]. This depends solely on the practice of the referring oncologist, although we would usually recommend consideration of adjuvant chemotherapy in biphasic disease or pathological proven nodal disease [15]. Given the lack of evidence regarding the optimal timing of chemotherapy as part of multimodality therapy with radical surgery with EPD, there are no standardized indications for adjuvant therapy at present in the UK.

Statistical analysis

SPSS version 20 statistical software package was used for analysis. Continuous data were analysed using the Mann–Whitney U-test, as all continuous dependent variables in this study were not normally distributed. A histogram was produced for all continuous variables and normality checked visually. Also, a Shapiro–Wilk test was performed and Q–Q plots were created to ensure whether data were normally distributed. Categorical data were compared between the 2 groups (<70 years or ≥70 years) using χ² test, or in cases when one or more of the cells had an expected frequency of 5 or less, the Fisher’s exact test was used. A P-value of <0.05 was considered to be statistically significant. Progression-free and overall survival analyses were estimated using the Kaplan–Meier method with log-rank test to compare for differences between the groups. The multivariable model was created using a Cox regression analysis. Variables with a P-value of < 0.1 were included in the model.

RESULTS

Patients

Three hundred patients were included in this study: 218 (72.7%) patients in the <70 years group and 82 (27.3%) patients in the ≥70 years group (Table 1). Median age of the entire cohort was 65 (range 42–81) years. In both groups, the majority of patients were male (83%), and the majority had epithelioid subtype (<70 years 77.5%, ≥70 years 76.8%; P = 0.95). Most patients were pathological T Stage 3 (<70 years 52.3%, ≥70 years 47.6%; P = 0.18) and a majority had lymph node metastases at operation (<70 years 59.2%, ≥70 years 59.8%; P = 1.0).

More patients in the <70 years group received neoadjuvant chemotherapy (<70 years 21.2%, ≥70 years 11.0%; P = 0.045). There was no statistically significant difference between the groups in terms of preoperative anaemia, thrombocytosis, lymphocytosis or raised creatinine levels (Table 1). Both groups had a similar median preoperative forced expiratory volume in 1 s [<70 years 71.5 (44–123)%, ≥70 years 70 (40–127)%; P = 0.96]. More patients were in performance Status 0 in the <70 years group, although this did not reach statistical significance (<70 years 61.1%, ≥70 years 50.6%; P = 0.28).

Over this same time period, 134 patients underwent EPP. Median age of these patients was 57 (range 14–70) years. Similarly, to those undergoing EPD, the majority had epithelioid subtype (72.4%) and were pathological T Stage 3 (47.8%). Despite routine mediastinoscopy prior to EPP, 53.0% of patients were either N1 or N2 positive at operation. Median overall survival of all these patients, including all cell types and pathological stages, was 12.9 months.

Perioperative course

There was no difference in median in-hospital length of stay between the 2 groups [<70 years 12 (5–70) days, ≥70 years 14 (6–93) days; P = 0.15].

More patients in the <70 years group received adjuvant chemotherapy (<70 years 45.4%, ≥70 years 29.3%; P = 0.04). Those with a preoperative performance status of 2 (3 patients) did not receive adjuvant chemotherapy. Of the patients with a performance status of 1, 37.0% (34) patients received adjuvant therapy compared with 45.9% (51) patients of those with a performance status of 0 (P = 0.086).

Data regarding postoperative complications were available for 287 patients (<70 years 209 patients, ≥70 years 78 patients) (Table 2). There was a trend towards an increased incidence of postoperative atrial fibrillation in the older group (14.4 vs 24.4%, P = 0.05). There was a higher proportion of patients requiring admission to the ICU postoperatively in the older group (6.2 vs 16.7% P = 0.01). In the majority of cases, this was for inotropic support (<70 years 76.9%, ≥70 years 53.8%). Reintubation was required in 1 patient in the <70 years group (7.7%) and 3 patients in the ≥70 years group (23.1%). The remaining 5 patients suffered major postoperative complications such as sepsis with or without multiorgan failure. Despite this, there was no difference in terms of in-hospital, 30-day or 90-day mortality between the 2 groups (Table 2).

Survival outcomes

There was no difference in the median overall survival between the 2 groups: <70 years 14.0 months, ≥70 years 10.3 months; P = 0.29 [hazard ratio (HR) 1.17, 95% confidence interval (CI) 0.872–1.580] (Fig. 1), i.e. age was not found to be a predictive factor for overall survival in patients undergoing EPD. There was also no difference in median progression-free survival: <70 years 12.6 months, ≥70 years 14.5 months; P = 0.68 (HR 1.09, 95% CI 0.733–1.616). During the entire follow-up period, 222 (74%) patients died.

In those with epithelioid disease, there was no significant difference between the age groups in terms of median overall survival: <70 years 17.7 months; ≥70 years 12.5 months; P = 0.14 (HR 1.30, 95% CI 0.917–1.830). The same was found for those with non-epithelioid disease: <70 years 7.8 months, ≥70 years 6.5 months; P = 0.674 (HR 0.88, 95% CI 0.489–1.588). There was no difference in median overall survival between the age groups in those without nodal spread: <70 years 17.8 months, ≥70 years 16.0 months; P = 0.87 (HR 1.05, 95% CI 0.578–1.513). However, in those with nodal disease at operation, elderly patients had a poorer median overall survival: <70 years 13.0 months, ≥70 years 9.1 months; P = 0.05 (HR 1.46, 95% CI 1.002–2.140) (Fig. 2). In patients with non-bulky disease (pT Stage 1 or 2), there was no median overall survival difference between the groups: <70 years 16.2 months, ≥70 years 16.0 months; P = 0.83 (HR 0.94, 95% CI 0.554–1.604). In those with bulky disease (pT Stage 3 or 4), there was a trend towards a poorer median overall survival in older patients: <70 years 13.2 months, ≥70 years 9.2 months; P = 0.06 (HR 1.41, 95% CI 0.984–2.028) (Fig. 3).

In patients with non-epithelioid disease and nodal spread (n = 28), there was a poorer median overall survival if they were ≥70 years of age (6.7 vs 3.8 months, P = 0.04) (HR 2.13, 95% CI 1.001–5.320) (Fig. 4).

One-year survival was 47% in the younger (<70 years) group and 38% in the older (≥70 years) group. Three-year survival was 8.7% in the younger and 11% in the older groups. Five-year survival was 4.1% in the younger group and 0% in the older group.

On multivariable analysis of factors known to affect outcome in patients with MPM [16], non-epithelioid histological subtype (HR 2.90, 95% CI 1.896–4.437; P < 0.001), lymph node spread (HR 1.64, 95% CI 1.071–1.918; P < 0.01), preoperative anaemia (HR 1.59, 95% CI 1.121–2.255; P = 0.011), not receiving adjuvant chemotherapy (HR 1.93, 95% CI 1.390–2.688; P < 0.001) and immediate postoperative ICU admission (21 of 25 patients admitted to intensive care) (HR 3.13, 95% CI 1.546–6.322; P = 0.01) were all independent prognostic factors for overall survival (see Supplementary Material, Table S1).

DISCUSSION

We have shown that it is possible to safely perform an EPD on patients older than 70 years, with no detriment to short- or long-term outcomes. Age itself was not a prognostic factor for poor overall survival outcomes, and as such, it should not be used in isolation as a selection factor when determining suitability for this type of surgery.

EPD has been shown to be a less morbid operation than EPP, with a 27.9% morbidity rate compared with 62% with EPP [17]. Moving to performing EPD as the operation of choice for radical resection has allowed us to operate on older patients, with more comorbidities and surgical risk factors [11]. Although the selection criteria for EPD are less strict than for EPP, it still remains an operation with considerable morbidity and mortality risk, and good patient selection is paramount [7, 8, 18].

The majority of outcome data for both operative procedures come from retrospective institutional reports with differing patient selection criteria and policies regarding neoadjuvant or adjuvant therapy provision and as such are difficult to compare. Similarly, many do not report stage or histological subtype-specific survival, which can be influential when comparing outcomes. Several retrospective studies have shown that EPD can be performed with good long-term oncological and overall survival outcomes [4, 6]. One single-institution prospective study of multimodality therapy, with EPD as the surgical approach, showed a median overall survival of 30 months [5]. In the recent systematic review of multimodality therapy in MPM from Cao et al. [7], outcomes of EPP and EPD were compared. Median overall survival ranged between 13 and 29 months for EPD and 12 and 22 months for EPP [7]. This study concluded that EPP and EPD can be performed safely with good long-term oncological outcomes, in selected patients, in specialized centres.

The age of patients presenting with MPM is increasing in the UK, with the median age currently 73 years [9]. This increasing age at presentation is probably due to the changing patterns and levels of exposure to asbestos over time; e.g. patients having lower level exposure over longer periods of time as usage of the material varied over the last 3 decades or exposure via relatives working with the material as opposed to direct contact. There was also an increase in the usage of amosite asbestos prior to the ban towards the end of the last century, and it is thought that the latency period is increased following exposure to this particular type of asbestos, giving a later age at presentation.

This study has shown that we are able to perform macroscopic complete resection, in the form of EPD, in patients older than 70 years, without affecting long-term overall and progression-free survival outcomes. However, although it did not reach statistical significance, it should be noted that there were no patients older than 70 years alive at 5 years, when compared with 4.1% in the younger than 70 years age group.

Those patients older than 70 years had a more complex postoperative course, with a higher proportion developing atrial fibrillation and a higher proportion requiring admission to the ICU. In the majority of cases, admission to ICU was for short-term (1–2 days) inotropic support. Despite this, there was no difference in in-hospital length of stay or in in-hospital, 30-day or 90-day mortality rates between the 2 groups of patients. Those selected for EPD, either ≥70 years or < 70 years of age, must have sufficient physiological reserve to tolerate the initial respiratory insult and the potential complications of this major operation, and as such, we must ensure that we continue to be highly selective when determining a patient’s fitness for surgery and include objective measures of physiological fitness where possible.

Admission to ICU itself was found to be a significant predictor of poor overall survival. In particular, the need for inotropic support must be avoided where possible through good preoperative physiological optimization and accurate intraoperative fluid management, especially in the elderly population.

We found there to be a significantly higher proportion of patients in the younger (<70 years) age group receiving neoadjuvant and/or adjuvant chemotherapy. It is unclear as to why there was a higher proportion receiving neoadjuvant therapy; this may be due to patient preference or possibly oncologist choice in cases where they felt that elderly patients may be rendered unfit for surgery following neoadjuvant chemotherapy or if they are seen to be too old or have too many comorbidities to tolerate chemotherapy upfront. The decision as to whether to give adjuvant chemotherapy is made by the referring oncologist and is based on patient fitness, as well as oncologist standard practice, and, in some cases, histological subtype or lymph node spread [15]. The only factor directly affecting adjuvant chemotherapy provision was early space infection, and adjuvant chemotherapy was not offered to these patients with ongoing infection. Given that fitness for radical surgery does not necessarily equate to fitness for adjuvant chemotherapy, an additional assessment for adjuvant chemotherapy must be performed preoperatively. For example, in our centre, chronic kidney disease does not necessarily preclude a patient from undergoing EPD; however, this may lead to a patient not receiving adjuvant chemotherapy. Given the provision of adjuvant chemotherapy is a prognostic factor for survival, and is also lower in the elderly group, we must ensure that patients are going to remain fit for this treatment following EPD before we agree to perform this procedure. As opposed to the standard practice of a surgeon solely determining fitness for surgical resection, it may be appropriate for further discussion with our oncology colleagues after initial surgical assessment to determine whether patients are likely to be suitable for chemotherapy treatment following resection. In cases where patients are of borderline fitness, it may be more suitable to give neoadjuvant chemotherapy, although there is limited evidence to show a benefit of this treatment strategy prior to EPD [19]. We must ensure that patients, particularly those of borderline fitness or of advanced age, are not rendered unfit for adjuvant therapy due to the morbidity of this major procedure. Although not statistically significant, there was a difference in the proportion of patients receiving adjuvant therapy depending on preoperative performance status, which may represent their physiological ability to tolerate the morbidity associated with EPD.

Limitations

The main limitation of this study is its retrospective nature. It is likely that those older than 70 years were strictly selected and do not represent the entire cohort of patients in this age group with the disease. However, this represents the problem we face as clinicians; we must select our patients carefully for this operation which is not without mortality and morbidity risk. The MARS2 trial: A Feasibility Study Comparing (Extended) Pleurectomy Decortication Versus no Pleurectomy Decortication in Patients With Malignant Pleural Mesothelioma (MARS2) (NCT02040272) is currently recruiting in the UK and seeks to determine the role of EPD in the treatment of MPM. The Phase III trial following on from this feasibility trial will include patients older than 70 years who will be randomized to chemotherapy plus surgery or chemotherapy alone and should allow us to determine whether there is a true benefit to pursuing this radical treatment strategy in the elderly population. Another limitation is the variability in the provision of neoadjuvant or adjuvant chemotherapy. As a tertiary referral centre, we receive referrals from all over the UK, and the practice of oncologists in different units varies. This could result in biased results with regard to the effect of these therapies, alongside EPD, on survival outcomes.

It has been shown that the quality of life following EPD is more markedly improved in patients who were symptomatic prior to their operation compared with those who were in performance Status 0, without detriment to the pulmonary function [20]. It may be useful to determine the changes in the quality of life in elderly and younger patients, as it may be possible that older patients with more comorbidities have in fact more to gain from undergoing EPD than younger, possibly less symptomatic, patients. Quality-of-life assessment is an important aspect of the MARS2 study, so we expect to gain an insight into the benefits or otherwise with radical surgery from this analysis.

CONCLUSION

Although age itself should not be a criteria for selection for radical surgery with EPD, we must ensure that we do not render our patients unfit for adjuvant chemotherapy and that we avoid the need for postoperative intensive care admission where possible. We must remain highly selective, while not excluding patients from undergoing EPD based solely on their advanced age. However, in cases of bulky disease or possible nodal disease, caution should be employed when selecting patients older than 70 years, given their potential poor overall survival following EPD.

SUPPLEMENTARY MATERIAL

Supplementary material is available at ICVTS online.

Conflict of interest: none declared.

REFERENCES

Author notes