Comparison of treatment outcomes in propensity score matched primary lung cancer patients offered anatomical lung resection or oncological treatment following a thoracic high risk multidisciplinary team meeting

Introduction

Multidisciplinary team (MDT) meetings are mainly utilized in lung cancer treatment pathways. However, they are often characterized by over-filled lists, incomplete pre-operative assessments, inadequate staffing and limited time. Consequently, patients may not obtain the comprehensive in-depth discussions they require (1-3). Furthermore, most lung cancer MDTs make decisions without taking into consideration the opinion of other important parties, such as anaesthesia and physiotherapy, which are vital, particularly for the peri-operative management of high-risk patients (1-3). However, a standardised definition of ‘high-risk’ in thoracic surgery, has yet to be defined and as such this is usually determined according to local, and highly variable, protocols (4-6).

The current standard is that oncological treatment for patients deemed as high-risk is perceived as a safer choice than surgery, with surgeons being reluctant to offer surgery to those patients, under the fear of adverse outcomes (7). In order to overcome this, High-risk multidisciplinary team (HRMDT) are increasingly implemented (3-5,8,9). To date however, there has been no direct comparison of surgery and non-surgical treatment of high-risk lung cancer patients at all stages of lung cancer.

The aim of this study was to compare the survival outcomes between propensity matched primary lung cancer patients who received anatomical lung resection surgery with those who received oncological treatment, following discussion at a HRMDT. Multiple other outcomes related to treatment were also compared between the 2 groups. We present this article in accordance with the STROBE reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-24-184/rc).

Methods

All patients treated for non-small cell lung cancer (NSCLC) after discussion at our local HRMDT were included in this study and reviewed in a retrospective manner. The study period was between 2019–2023. Patients were divided into a surgical and an oncology treatment group. The surgical group involved only those who had received anatomical lung resections, excluding wedge resections, in order to involve surgery with maximum curative probability. The oncology group included patients who were excluded from surgery and were referred to oncology for treatment with curative intent. Propensity score matching was performed to create similar patient treatment groups. Data on overall mortality/survival and other outcomes including disease free survival were compared between the two matched groups.

Our institution’s HRMDT was introduced in May 2019 and is attended by 5 thoracic surgeons, 1 intensive care consultant and 1 thoracic anaesthetist, allowing for pre-operative optimisation and planning for high dependency unit (HDU) support. A member from the physiotherapy team, the Thoracic specialist nurses and the ward sisters also participate, allowing for planning of needs in the immediate post-operative period. The meeting is held on a fortnightly basis. The HRMDT is independent from the main lung MDT. After initial review by the surgeon in clinic, patients are discussed at the HRMDT according to the referral criteria. The criteria/indications for referral, agreed amongst the members of the HRMDT, are summarized in Table 1, and pertain to the clinical condition of the patients. The main reason of referral is captured as in many cases patients had more than 1 reasons for referral. The cases discussed refer only to surgically resectable cases or cases in which surgical treatment was decided from the lung MDT. During the discussion at the HRMDT, consensus is sought amongst all consultants present. Therefore, the operating surgeon always follows the HRMDT recommendation. After discussion, a decision is reached as to whether or not surgery will be offered to the patient. If surgery is not offered, the patient is then referred to the oncology service at which it is decided what non-surgical treatment is appropriate {i.e., chemotherapy +/− radiotherapy [and stereotactic beam radiotherapy (SBRT)] +/− immunotherapy}. Patients discussed at the HRMDT and were offered surgery after neo-adjuvant treatment were still included into the surgical group because this was the main reason they were referred to the HRMDT. If surgery is offered to the patient, the operation is scheduled such that another consultant surgeon is available on the day of the high-risk procedure. Surgical patients then follow the appropriate adjuvant treatment pathways (i.e., chemotherapy, immunotherapy, etc.), as needed according to staging.

Patient data

Data was retrospectively collected about all eligible patients including non-identifiable demographic details (age, gender), co-morbidities, body mass index (BMI), and lung function test (LFT) results, including forced expiratory volume in 1 sec (FEV1 in %), the transfer capacity of the lung (TLCO in %). General preoperative fitness was assessed using cardiopulmonary exercise test (CPET), the American Society of Anesthesiologists (ASA) physical status classification (ASA grade), and the performance status (PS) as per the Eastern Cooperative Oncology Group (ECOG) (10). Additional data collected included the pathological stage of the cancer on histology, the reason for discussion at the HRMDT (as per Table 1), the overall length of stay (LOS) (including duration of in-hospital stay during index admission as well as duration of any re-admissions). The morbidity captured included those related to treatment. For surgical patients, post-operative complications were recorded, including prolonged air leak (air leak >5 days), postoperative bleeding, and redo surgery. For oncological patients, post-treatment complications related to oncological treatment were included, such as pneumonitis, fibrosis and stroke. Furthermore, for all patients, other general complications not related to treatment were included, such as respiratory infection/failure, cardiac, and renal failure. Death related to disease progression, or otherwise, in-hospital, at 30-days postoperative, and at 90-days postoperative was recorded, as well survival until the end of the study. Recurrence was identified at the time of the diagnosis. If patients were discussed more than once, data from the final discussion only was included in the final analysis to omit duplicates. There were no missing data in all variables captured.

Grouping of patients

On completion of data collection, propensity score matching was conducted in a 1:1 matching manner with the “nearest neighbour” technique, allowing for 0.5 matching tolerance. Two groups were created: the first group included 63 cases of patients who received surgery (anatomical lung resection) as main treatment for their NSCLC after discussion at the HRMDT and a second one which included 63 cases of patients who received oncological treatment (i.e., non-surgical) as the main treatment for their cancer. Post matching investigations were performed to designate similarity of the groups.

Approval

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by Information Governance service board of Norfolk and Norwich University Hospital (No. 05-20-001) as a service evaluation project assessing the implementation of the HRMDT implementation within the institution. As all patients had previously provided consent for use of their routine data, no additional informed consent was required.

Statistical analysis

Statistical analysis was performed with IBM SPSS Statistics for Macintosh, Version 25.0. Armonk, New York: IBM Corp. All numerical data were investigated for normality of distribution with Shapiro-Wilk’s test (normally distributed when P>0.05) and Q-Q plots (acceptable figures for normality of distribution) and were presented as mean ± standard deviation (SD). Skewed data were presented as median (range). Categorical data were presented as number of observations and percentages.

Statistical significance was determined with Student’s t-test for normally distributed data, or Mann-Whitney for data that was not normally distributed or lacked homogeneity of variance (Levene’s test P<0.05). Chi-square and Fisher’s Exact tests were used for comparison of categorical data. The statistical significance level was set to a P value <0.05. Log-rank was used to investigate differences in Kaplan-Meier survival curves and disease-free survival between the groups. This was done in the overall cohort and by NSCLC stage. Deaths not related to cancer were included in this analysis, as deaths related to co-morbidities were perceived important to be included in the investigations, in order to truly capture the differences in survival outcomes of high-risk patients. Survival was calculated from the day of the HRMDT discussion. The last date investigated for survival before submission was 1^st March 2024. The last date for disease-free survival was the date of the MDT verifying recurrence or metastasis (or 1^st March 2024 if no recurrence or metastasis was documented). There were no patients lost from follow up due to strict cancer pathway. Cox’s regression was conducted to identify if surgery in these specific patients is perceived as a prognostic factor of death by time.

Results

Within the study period, 168 patients were discussed at the HRMDT for treatment of NSCLC and hence were included in the analysis. The overall mean age was 72.3±2.2 years and 73 (43.4%) were males. 103 (61.3%) patients received decision for surgery and 65 (38.7%) for oncological treatment. The main reason of referral for discussion at the HRMDT was poor LFTs/CPET (92 patients, 54.8%), reduced ASA because of important co-morbidities (33 patients, 19%), important frailty and PS ≤2 (28 patients, 16.7%), age ≥85 years (10 patients, 5.9%) and BMI issues (6 patients, 3.6%).

Comparison between these 2 groups (Table 2) showed differences in age, gender, NSCLC stage and important co-morbidities (as per Table 1). These variables were therefore included in the propensity matching process.

After propensity matching (Table 3), 2 groups were created, with 63 matched patients in each group. After the matching process, variables that were previously different between the unmatched groups (age, gender, NSCLC stage and co-morbidities) were found to be similar between the 2 matched groups. PS, LFTs, reason for referral and smoking status across the matched groups were also similar (Table 3).

Post-treatment outcome comparison of the matched surgery and oncology groups (Table 4) showed that surgical treatment resulted in a longer LOS (median 5 vs. 4 days, P=0.03) and more unexpected admissions to HDU [6 (9.5%) vs. 0 (0%), P=0.01], but less incidence of recurrence [9 (14.3%) vs. 29 (46%), P<0.001] and reduced overall deaths [6 (9.5%) vs. 24 (38.1%), P<0.001, Table 4]. A similar number of non-cancer deaths [7 (11.1%) vs. 6 (9.5%)], in-hospital, 30- or 90-day mortalities was recorded between the 2 groups (Table 4).

Overall complications from treatment had the tendency to be higher in the surgical compared to oncology group [34 (54%) vs. 26 (41.3%), P=0.09], involving mainly pneumonia [27 (42.9%)] with lung consolidation [needing further interventions such as bronchoscopy in 3 cases (4.8%), tracheostomy in 2 (3.2%) etc.], atrial fibrillation [12 (19.05%)] and pleural cavity infection [8 (12.7%)], in the surgical group. Common complications in the oncological group included pneumonitis [8 (12.7%)], lung fibrosis with progressive breathlessness [3 (4.8%)] and organ failure [2 (3.2%), heart failure and acute kidney injury].

Overall survival and overall disease-free survival

Two patients died within 30 days from surgery due to surgery related issues. Three patients died within 90 days in the oncology group because of massive stroke and rapid disease progression while only 1 in the surgical group, after prolonged in-hospital stay with lung infection. The overall survival was longer in the surgical group (Figure 1A, log-rank =13.944, P<0.001). Similarly, the disease-free survival was longer in the surgery group (Figure 1B, log-rank =16.471, P<0.001). None of the patients were lost to follow up, which was performed with a median of 29 months (range, 1–58 months).

Figure 1 Overall survival curve (A) and disease-free survival (B) of 126 matched patients according to treatment (surgery versus oncology treatment).

Analysis using Cox’s regression showed that surgical treatment was associated with a statistically significant reduction in the risk of death over time [B=−1.766, confidence interval (CI): 0.064–0.457, P<0.001].

Survival and disease-free survival per NSCLC stage subgroup analysis

For patients with stage 1 disease, survival was longer in the surgical group than in the oncology group (log-rank =3.858, P=0.046), as was the disease-free survival (log-rank =8.570, P=0.003). Similarly, patients with stage II and III disease survived longer or had the tendency to do so after surgery than oncological treatment (log-rank =3.913, P=0.048 and log-rank =3.104, P=0.07, respectively). The disease-free survival was similarly longer or had the tendency to do so in stage II and III surgical patients when compared with their oncology counterparts (log-rank =5.117, P=0.02 and log-rank =2.036, P=0.054, respectively).

Discussion

The main finding of the present study was that surgery for NSCLC in high-risk patients improves both overall and the disease-free survival when compared to oncology group. Similarly, longer survival and disease-free survival was observed after surgical treatment at all lung cancer stages (tendency to be important in stage III). Not offering surgery was a predictor of death by time and surgery did not induce increase of in-hospital or 30-day deaths due to surgical complications. Despite this, high-risk patients offered surgery, experienced more overall complications, longer LOS and more unplanned admissions to a critical care setting. Accepting the postoperative shortcomings, surgery should not be precluded from the treatment of high-risk NSCLC patients, provided the input of HRMDT is ensured.

MDTs have been well established for decades, allowing for a collaborative approach between specialties to streamline the pathway of patients diagnosed with lung cancer (2,11). The HRMDT follows a similar set-up and, in our study, its implementation to the practice of our busy service provided support for the clinicians who treat such high-risk patients. Our results show that patients were more streamlined towards a resection, possibly because surgeons were keener in offering surgery to those patients, feeling supported and protected through an MDT approach decision.

The HRMDTs are not routinely used and as such, there are no clear indications as to who should get discussed. The use of Thoracoscore in the past gained interest in the stratification of risk in thoracic surgery patients but it has been shown to have low clinical correlation with short-term mortality (12,13). In this study, our criteria of referral were agreed with all HRMDT members. This event could raise bias in our study but with these common criteria used for both groups, surgical treatment was proven equally safe with the non-surgical one.

An important finding of this study was that the overall survival was longer in the surgical group, with surgical treatment outlined as a statistically significant factor in reducing the risk of death over time. Our study suggests that surgery, in this cohort of high-risk patients, can offer a survival benefit over patients who are treated solely with oncological measures. Furthermore, our study found longer overall and per stage disease-free survival with less recurrences in the surgical group. This finding needs further exploration, in terms of other variables, such as R1 resections and the number of operations required. Also, because the number of cases was small within stage groups the finding of better survival/disease-free-survival needs to be cautiously interpreted.

Oncological treatment is often considered to be safer than surgery (9). Higher risk patients, consequently, are easier shifted towards oncology for their NSCLC treatment. In other studies, surgery in high-risk patients is by definition perceived as “troublesome” and its safety is questioned when compared with outcomes from surgery performed in “normal” or “low-risk” patients (3,4). However, a direct comparison between the perceived high-risk surgery versus the safer oncological treatment, would be more appropriate. Our results show that, even in advanced stage patients, surgery is safe and more beneficial in terms of survival and cancer recurrence than with oncological treatment. And, although, previous studies on high-risk patients mainly involved early-stage cancers, our study shows that even in progressed stage III patients, surgery still remains a safe and successful choice (4). Lastly, very few patients died because of surgical complications and surgery did not inflict more non-cancer related deaths when compared with oncological treatment. These findings augment the previously mentioned notion, that surgery in “high-risk” patients is safe and it does not induce additional mortality because of the procedure. This finding, as described before, needs to be considered in a HRMDT framework and preparation for the resection.

Another finding of this study was that respiratory compromise, i.e., poor LFTs, should not be perceived as the primary reason for patients to be perceived as “high-risk”. In this study LFTs were found to be similar before and after the matching process, proving that poor LFTs did not shift the decision towards oncological treatment. More importantly, frailty/PS, age, many co-morbidities and BMI propelled surgeons to refer to the HRMDT. As such, the addition of a physician would have, perhaps, aided in providing a balance of medical co-morbidities optimization against surgical risk (14). Increasing scrutiny on the profession almost dictates that decisions should be made as a unit, with a thorough and holistic approach (15). Equally, an oncologist is not involved in the discussions during the HRMDT as no oncological treatment is decided during that meeting. As a surgical-focussed MDT, by not involving respiratory physicians or oncologists we avoided recreating another lung MDT which could involve long oncological discussions, hence maximized time efficiency with decisions streamlined only to those who are appropriate for surgery.

Surgery in this study involved lung resections from sublobar ones to pneumonectomies. It is known that more extended resections might carry higher morbidity and mortality risk, although in appropriately selected patients such resections may actually have a beneficial effect, such as the long volume reduction effect in patients with emphysema (16,17). Similarly, many cases involved robotic resections and hence segmentectomies may be more easily performed, sparing lung parenchyma (18,19). This is a possible limitation of the study as the results were not adjusted for the type of resection received. Similarly, the oncological group involves a range of treatments including chemotherapy, stereotactic radiotherapy, classic radiotherapy and immunotherapy, following the widely accepted treatment guidelines (2,20), with oncological results also not adjusted for this.

Apart from the above, this study has further limitations. Firstly, it is a single institute retrospective study involving a small number of patients. Secondly, bias could be coming from surgeons who could impose their opinion by subjectively presenting the case to the MDT favouring surgery or not. Equally, the patients were not reviewed by another consultant which enhances the chances of bias. Lastly, no quality-of-life data post treatment amongst groups have been obtained. The findings, therefore, of this study need to be carefully considered. Further studies are warranted in order to clarify the best treatment option in high-risk NSCLC patients.

Conclusions

In conclusion, surgery as treatment for high-risk NSCLC patients overall is shown to be equally safe to oncological treatment and to provide longer overall and disease-free survival. However, post-treatment outcomes were worse with surgery and hence all these need to be considered when deciding the best treatment to be offered to these patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://atm.amegroups.com/article/view/10.21037/atm-24-184/rc

Data Sharing Statement: Available at https://atm.amegroups.com/article/view/10.21037/atm-24-184/dss

Peer Review File: Available at https://atm.amegroups.com/article/view/10.21037/atm-24-184/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-24-184/coif). V.K. serves as a proctor for Intuitive and an unpaid editorial board member of Annals of Translational Medicine from December 2025 to November 2027. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by Information Governance service board of Norfolk and Norwich University Hospital (No. 05-20-001) as a service evaluation project assessing the implementation of the HRMDT implementation within the institution. As all patients had previously provided consent for use of their routine data, no additional informed consent was required.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Zasada M, Harris J, Groothuizen J, et al. Investigating the efficiency of lung multi-disciplinary team meetings-A mixed methods study of eight lung multi-disciplinary teams. Cancer Med 2023;12:9999-10007. [Crossref] [PubMed]
2. Lim E, Baldwin D, Beckles M, et al. Guidelines on the radical management of patients with lung cancer. Thorax 2010;65:iii1-27. [Crossref] [PubMed]
3. Nwaejike N, Elbur E, Malagon I, et al. Is there a role for the high-risk multidisciplinary team meeting in thoracic surgery? Interact Cardiovasc Thorac Surg 2016;22:397-400. [Crossref] [PubMed]
4. Sancheti MS, Melvan JN, Medbery RL, et al. Outcomes After Surgery in High-Risk Patients With Early Stage Lung Cancer. Ann Thorac Surg 2016;101:1043-50; Discussion 1051. [Crossref] [PubMed]
5. Puri V, Crabtree TD, Bell JM, et al. National cooperative group trials of "high-risk" patients with lung cancer: are they truly "high-risk"? Ann Thorac Surg 2014;97:1678-83; discussion 1683-5. [Crossref] [PubMed]
6. Taylor M, Szafron B, Martin GP, et al. External validation of six existing multivariable clinical prediction models for short-term mortality in patients undergoing lung resection. Eur J Cardiothorac Surg 2021;59:1030-6. [Crossref] [PubMed]
7. Pepper JR, Coonar AS. High-risk surgery: the courage to fail. J R Soc Med 2015;108:44-6. [Crossref] [PubMed]
8. Fernando HC, Timmerman R. American College of Surgeons Oncology Group Z4099/Radiation Therapy Oncology Group 1021: a randomized study of sublobar resection compared with stereotactic body radiotherapy for high-risk stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 2012;144:S35-8. [Crossref] [PubMed]
9. Crabtree T, Puri V, Timmerman R, et al. Treatment of stage I lung cancer in high-risk and inoperable patients: comparison of prospective clinical trials using stereotactic body radiotherapy (RTOG 0236), sublobar resection (ACOSOG Z4032), and radiofrequency ablation (ACOSOG Z4033). J Thorac Cardiovasc Surg 2013;145:692-9. [Crossref] [PubMed]
10. Performance Status ScaleECOGECOG-ACRIN Cancer Research Group. Available online: https://ecog-acrin.org/resources/ecog-performance-status/
11. NHS Independent Cancer Taskforce. Achieving World-Class Cancer Outcomes: A Strategy for England 2015-2020. 2015. Available online: https://www.england.nhs.uk/wp-content/uploads/2016/10/cancer-one-year-on.pdf
12. Falcoz PE, Conti M, Brouchet L, et al. The Thoracic Surgery Scoring System (Thoracoscore): risk model for in-hospital death in 15,183 patients requiring thoracic surgery. J Thorac Cardiovasc Surg 2007;133:325-32. [Crossref] [PubMed]
13. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043-9. [Crossref] [PubMed]
14. Prabhakar CN, Fong KM, Peake MD, et al. The effectiveness of lung cancer MDT and the role of respiratory physicians. Respirology 2015;20:884-8. [Crossref] [PubMed]
15. Westaby S, De Silva R, Petrou M, et al. Surgeon-specific mortality data disguise wider failings in delivery of safe surgical services. Eur J Cardiothorac Surg 2015;47:341-5. [Crossref] [PubMed]
16. Korst RJ, Ginsberg RJ, Ailawadi M, et al. Lobectomy improves ventilatory function in selected patients with severe COPD. Ann Thorac Surg 1998;66:898-902. [Crossref] [PubMed]
17. Carretta A, Zannini P, Puglisi A, et al. Improvement of pulmonary function after lobectomy for non-small cell lung cancer in emphysematous patients. Eur J Cardiothorac Surg 1999;15:602-7. [Crossref] [PubMed]
18. Zirafa CC, Romano G, Sicolo E, et al. Robotic Surgery for Non-Small Cell Lung Cancer Treatment in High-Risk Patients. J Clin Med 2021;10:4408. [Crossref] [PubMed]
19. Kneuertz PJ, Zhao J, D'Souza DM, et al. National Trends and Outcomes of Segmentectomy in the Society of Thoracic Surgeons Database. Ann Thorac Surg 2022;113:1361-9. [Crossref] [PubMed]
20. National Institute of Health and Care Excellence. Lung cancer: diagnosis and management. NG122. NICE; 2019. Available online: https://www.nice.org.uk/guidance/ng122