Non-small cell lung cancer (NSCLC) is a leading cause of mortality worldwide, accounting for 90% of lung cancer cases. Recent advances in imaging detection with thin-slice computed tomography (CT) have allowed early identification and timely treatment of this condition. In early-stage NSCLC, lobectomy with lymph node dissection has been the gold standard approach, offering cure and 5-year survival rates of up to 95% [1,2]. However, this surgical technique may not be the best option in patients with serious preoperative comorbidities, poor lung function, or frailty, which may contribute to perioperative complications and poorer patient survival. Subpulmonary or wedge resection is an attractive approach that has gained surgical ground due to several advantages regarding the preservation of lung tissue, fewer complications, and minimally invasive approaches [3-5]. Recent studies have shown comparable survival rates between wedge resection and lobectomy in early-stage NSCLC smaller than 2 cm; however, debate continues due to the uncertainty of the results and limited information from clinical implementation [6-8]. Therefore, our study aimed to explore whether each technique had equivalent results regarding survival and radiologic-pathologic variables in a novel Peruvian registry of patients treated for NSCLC.

Design and sample size

This was an analytical, longitudinal, and retrospective study. All patients with early-stage NSCLC who underwent surgical lung resection with lymph node dissection by our thoracic surgery service during the period 2000–2020 were eligible for inclusion. Early-stage NSCLC included stages I–II, meaning that the lung tumor had grown and might have spread to the lymph nodes but had not yet spread to organs beyond the lungs. We used CT scans to assess the lung cancer staging, and histopathology and node findings in the surgical approach contributed to the final clinical staging. Edgardo Rebagliati Martins National Hospital is among the leading centers nationally in the surgical treatment of lung oncological pathology, and its patient base exceeds 12% of the national affiliated population. All patients who had undergone a lobectomy or wedge resection together with a lymph node dissection (≥1 lymph node removed), and who had complete preoperative, intraoperative, and postoperative records up to 5 years after hospital discharge, were enrolled in the study.

Lung resection and lymph node dissection

For lobectomy and wedge resection, the traditional method was the posterolateral thoracotomy, but its utilization has diminished over the years against approaches that use video-assisted thoracic surgery (VATS) and robot-assisted thoracic surgery. Nevertheless, it remains the preferred approach in challenging scenarios, such as lung cancer with centrally located tumors, tumors larger than 6 cm, and dense adhesions resulting from inflammatory processes. The surgeon’s preference determines the sequence of the hilum approach and dissection, whether from posterior to anterior or anterior to posterior. Options for hilar division and repair include the use of staplers or hand-sewn techniques. Lymph node tissue was extracted to aid in cancer staging. A systematic examination of mediastinal lymph nodes was performed either as a systematic sampling, involving biopsies at each ipsilateral mediastinal station (stations 2R, 4R, 7, and 9R on the right, and stations 5, 6, 7, and 9L on the left), or through a formal mediastinal lymphadenectomy, where all associated nodes and soft tissue were excised between the anatomical landmarks.

Data collection and study variables

The main sources of information were each participant’s electronic medical record, operative reports, and outpatient evaluation form. All patients with the designation “postoperative lung resection” during the study period were enrolled. Their data were collected retrospectively and longitudinally and were organized according to the chronology of hospital care for 5 years following surgical treatment. The main variables were divided into 2 main groups by the type of lung resection (lobectomy or wedge resection) and the results of propensity-score matching (PSM). Baseline characteristics such as age, sex, body mass index, smoking, and comorbidities were considered. The primary comorbid diseases considered were chronic obstructive pulmonary disease (COPD), coronary heart disease, hypertension, and non-oncological lung conditions. The main criteria for discharge were the absence of severe pain, the patient’s ability to perform necessary tasks without associated issues (such as feeding and ambulation), a chest X-ray without pleural effusion, and relatively insignificant pneumothorax (<10%). In addition, the tomographic lesion size, presence of lymph nodes, VATS approach, histological typing, and intra- and postoperative complications were considered. The 5-year outcomes were overall survival (OS), disease-free survival (DFS), and locoregional-recurrence-free survival (LRFS). The combined analysis included the surgical resection margins (≤10 mm or >10 mm) and the presence of resected lymph nodes (<4 nodes or ≥4 nodes).

OS was defined as the time from the day of the surgical procedure to death from any cause, with censoring at the date of the last follow-up if death had not occurred within 5 years of the procedure. DFS was defined as the time from the day of the surgical procedure to either the first recurrence of NSCLC or death, with censoring at the date of the last follow-up if neither event had occurred or could be determined within 5 years of the procedure. LRFS was defined as the time from the day of the surgical procedure to either locoregional cancer recurrence, whether radiographically confirmed or pathologically sampled with the same histology as the index surgical procedure, or death.

The surgical team responsible for the patient’s procedure chose the type of lung resection with consideration for the patient’s baseline characteristics. Of particular note, the size of the tumor was prioritized over the consolidation-to-tumor ratio due to Edgardo Rebagliati Martins National Hospital protocol, and similar supporting evidence for both predictive variables. Likewise, we considered the data on the risk of perioperative mortality and each patient’s preference regarding the therapeutic options for their condition. To avoid bias, the patients in either group who had multiple affected nodules after surgery were sent for adjuvant therapy, such as chemotherapy, and followed similarly.

Statistical analysis

Categorical variables are presented as frequencies and percentages. Continuous variables are presented as mean± interquartile range, in some cases with standard deviation. To compare the lobectomy and wedge resection variables, PSM models were used. After matching, the imbalance was minimal with all standardized differences ≤10%, and these data were used to create inverse probability weights. The data were then re-weighted to ensure that the distribution of confounding factors was equal between the 2 groups under comparison. All relevant confounders were adjusted for, preserving the entire sample size and ensuring that the results of the study would be generalizable. The paired groups were compared through descriptive analyses (the unpaired Student t-test if the distribution was normal and the Mann-Whitney U test if it was not) and a Kaplan-Meier analysis, and multivariable Cox proportional hazard models were created for the 5-year OS, DFS, and LRFS outcomes. In all cases, the data collection, tabulation, and analysis were performed using Stata ver. 18.0 for MacOS (Stata Corp., College Station, TX, USA). Statistical findings were considered significant with a p-value of <0.05.

Ethical aspects

The study protocol was approved by the Ethics Committee of the Edgardo Rebagliati Martins National Hospital (IRB approval no., 1986-CTCV-12/24). The guidelines proposed by the Declaration of Helsinki were followed, data confidentiality was respected, and informed consent was not required due to the retrospective nature of the study.

The mean age of the participants was 68.56 years, and 64.3% were male. The most frequently occurring comorbidity was COPD in both the wedge resection group (44.1%) and the lobectomy group (33.7%). Based on CT, in both groups, the most common tumor size (31.4%) was 1.1 to 2.0 cm. Regarding histological features, the most frequently occurring cancer type among the participants was acinar adenocarcinoma (35.44%), followed by squamous cell carcinoma (29.02%), lepidic adenocarcinoma (12.14%), papillary adenocarcinoma (6.14%), micropapillary adenocarcinoma (6.14%), and solid adenocarcinoma (3.2%), and the majority had a moderate degree of differentiation (42.9%). VATS was used more often for wedge resection (70.3%) than for lobectomy (29.1%). The most common postoperative complication was pain persisting for >7 days, and it occurred more frequently in the lobectomy group than in the wedge resection group (58% versus 23%, p=0.034). The mean duration of hospitalization was shorter in the wedge resection group than in the lobectomy group (5.3 days versus 12.8 days, p=0.009) (Table 1). Regarding the 5-year and pre-PSM outcomes, no significant differences were found between the wedge resection and lobectomy groups regarding OS (84.3% versus 81.2%, p=0.091) or DFS (79.1% versus 74.1, p=0.07), although LRFS was found to be higher in the lobectomy group than in the wedge resection group (79.8% versus 91.1%, p<0.02) (Table 1). However, in the PSM analysis, LRFS was found to be more favorable in the wedge resection group with a statistically significant difference when the resection margin was >10 mm (90.9% versus 87.3%, p<0.048). Similarly, in patients with ≥4 resected lymph nodes, wedge resection showed a statistically significant advantage over lobectomy (82.8% versus 79.1%, p<0.011) (Figs. 1, 2).

Figure 1. Kaplan-Meier 5-year curve showing a comparison of the overall (A), disease-free (B), and locoregional-recurrence-free (C) survival rates of patients who experienced lobectomy or wedge resection for early-stage non-small cell lung cancer.

Figure 2. Highlights of outcomes and survival curves following wedge resection versus lobectomy for early-stage non-small cell lung cancer (NSCLC). PSM, propensity score-matched.

Table 1 . Comparison of outcomes and survival following wedge resection and lobectomy for early-stage non-small cell lung cancer.

Characteristic	Characteristics of patients before PSM analysis (n=2,570)				Characteristics of patients after PSM analysis (n=650)			Matched ASMD	Unmatched ASMD
	Wedge resection (n=1,845)	Lobectomy (n=725)	p-value		Wedge resection (n=325)	Lobectomy (n=325)	p-value
Age (yr)	75.98±15.46	69.98±15.46	0.201		68.17±13.15	69.23±15.34	0.090	0.084	0.747
Male	1,230 (66.6)	423 (58.3)	0.123		178 (54.7)	198 (60.9)	0.156	0.082	0.067
Body mass index ≥30 kg/m2	889 (48.2)	289 (39.8)	0.070		123 (37.8)	142 (43.6)	0.058	0.056	0.671
Smoking status			0.111				0.098	0.145	0.195
Current	867 (46.9)	328 (45.2)			139 (42.7)	127 (39.0)
Former	234 (12.6)	123 (16.9)			76 (23.38)	46 (14.15)
Never	744 (40.5)	274 (37.9)			110 (66.08)	152 (46.85)
Comorbidities								0.187	0.083
COPD	812 (44.1)	245 (33.7)	0.034		134 (41.2)	121(37.2)	0.091
Other (non-lung) cancer	578 (31.3)	189 (26.1)	0.093		98 (30.1)	75 (23.1)	0.054
Hypertension	379 (20.5)	98 (13.5)	0.058		48 (14.7)	59 (18.1)	0.089
Coronary artery disease	219 (11.8)	76 (10.4)	0.198		15 (4.6)	19 (6.6)	0.256
Radiographic workup
Clinical CT scan tumor size (cm)								0.076	0.143
0.0–1.0	445 (24.1)	134 (18.4)	0.099		87 (26.7)	56 (26.0)	0.198
1.1–2.0	589 (31.9)	217 (29.9)	0.075		108 (33.2)	98 (30.1)	0.094
2.1–3.0	314 (17.1)	141 (19.4)	0.092		25 (7.6)	67 (20.6)	0.055
3.1–4.0	199 (10.7)	123 (17.5)	0.061		85 (26.1)	87 (26.7)	0.114
>4.0	298 (16.15)	110 (14.1)	0.197		20 (6.1)	17 (4.9)	0.108
Multiple nodules (≥1)	768 (41.6)	290 (40.0)	0.195		111 (34.9)	98 (30.1)	0.080
Operative characteristics
VATS approach	1,298 (70.3)	211 (29.1)	0.031		154 (65.5)	121 (37.2)	0.032	0.081	0.634
Histology
Squamous cell carcinoma	478 (25.9)	268 (36.9)	0.055		113 (36.7)	129 (39.7)	0.153
Adenocarcinoma	1,256 (68.1)	457 (63.1)	0.083		212 (65.2)	196 (60.3)	0.089	0.190	0.627
Lepidic	234 (18.6)	78 (17.1)	0.104		34 (16.1)	21 (10.7)	0.157
Acinar	698 (55.5)	213 (46.6)	0.083		101 (47.6)	98 (50.0)	0.082
Papillary	134 (10.6)	24 (5.25)	0.052		57 (26.8)	26 (13.4)	0.062
Micropapillary	123 (8.6)	35 (8.9)	0.170		12 (5.6)	13 (6.6)	0.090
Solid	67 (4.7)	17 (4.5)	0.099		8 (4.1)	38 (19.3)	0.049
Differentiation								0.053	0.172
Mild	405 (21.9)	123 (16.9)	0.243		67 (28.1)	79 (24.3)	0.078
Moderate	789 (42.7)	313 (43.1)	0.132		134 (41.2)	156 (48.0)	0.099
Poor	652 (35.3)	289 (39.8)	0.099		124 (38.1)	90 (27.6)	0.153
Overall complications
Surgical bleeding >500 mL	156 (8.4)	95 (13.1)	0.055		78 (31.3)	99 (30.4)	0.189	0.091	0.446
Persistent pain >7 days	424 (23.0)	420 (58.0)	0.034		67 (28.1)	102 (31.3)	0.049	0.923	0.245
Hospital stay (day)	5.3	12.8	0.009		4.1	10.7	0.001	0.084	0.228
Persistent air leak	124 (6.7)	79 (10.8)	0.067		65 (26.5)	105 (33.1)	0.076	0.135	0.752
Outcomes at 5 years
Locoregional recurrence	198 (10.7)	134 (18.4)	0.070		22 (6.9)	17 (5.1)	0.195	0.186	0.852
OS	1,568 (85.6)	599 (82.7)	0.080		274 (84.3)	263 (81.2)	0.091	0.073	0.351
DFS	368 (80.1)	161 (77.9)	0.180		68 (79.1)	84 (74.1)	0.078	0.010	0.169
LRFS	373 (79.8)	65 (91.1)	0.020		59 (81.9)	55 (83.2)	0.055	0.434	0.046
5-year outcomes by wedge resection margin size and lymph node removal subgroups
OS, margin ≤10 mm	344 (81.4)	157 (78.4)	0.056		53 (80.1)	57 (82.3)	0.201	0.033	0.069
OS, margin >10 mm	344 (81.4)	114 (84.2)	0.109		17 (94.6)	25 (92.4)	0.184	0.765	0.074
DFS, margin ≤10 mm	503 (72.7)	218 (67.1)	0.091		45 (86.1)	46 (87.9)	0.276	0.864	0.319
DFS, margin >10 mm	503 (72.7)	190 (73.8)	0.079		88 (72.8)	74 (77.2)	0.088	0.131	0.064
LRFS, margin ≤10 mm	150 (91.9)	104 (85.6)	0.088		24 (92.6)	33 (89.7)	0.145	0.589	0.563
LRFS, margin >10 mm	150 (91.9)	52 (92.8)	0.123		29 (90.9)	41 (87.3)	0.048	0.045	0.011
LRFS, <4 lymph nodes removed	317 (82.8)	71 (90.8)	0.057		46 (85.7)	49 (84.9)	0.099	0.673	0.024
LRFS, ≥4 lymph nodes removed	408 (76.8)	133 (81.7)	0.062		56 (82.8)	68 (79.1)	0.011	0.463	0.584

Values are presented as mean±standard deviation or number (%)..

PSM, propensity-score matching; ASMD, absolute standardized mean difference; COPD, chronic obstructive pulmonary disease; CT, computed tomography; VATS, video-assisted thoracic surgery; OS, overall survival; DFS, disease-free survival; LRFS, locoregional-recurrence-free survival..

The present study found that the average age of the participants diagnosed with early-stage NSCLC was 74.2 years before PSM and 68.5 years after PSM. This agrees with the findings of Gomez Hernandez et al. [1], who identified the seventh decade of life as having the highest incidence of NSCLC, with an average of 65.1 to 67.9 years. Most of the patients included in this study were male, with an average of 64.3% before PSM and 57.1% after PSM, respectively. This is consistent with Bernard et al. [2] and Stamenovic et al. [3], who observed that NSCLC was diagnosed more frequently in male patients (averages of 64.2% and 67%, respectively), although the percentages were not significantly different between male and female patients. Li et al. [4] stated that 50.1% of patients with lung cancer had never used tobacco, a result similar to that of our study, where 46.49% of participants before PSM and 40.92% after PSM consumed tobacco. These results contradict those reported by Sakane et al. [5], who found that non-smoking patients represented only 35% of patients diagnosed with lung cancer.

Regarding the comorbidities most frequently present in patients suffering from NSCLC, our results indicated that it was COPD. Stamenovic et al. [3] concluded that presenting multiple comorbidities (Charlson score >11 points), including lung pathologies, was related to greater postoperative complications such as hemothorax, persistent pain that requires management with opioids, and pulmonary empyema [2]. Similarly, Li et al. [4] and Liu et al. [6] agreed that cardiorespiratory comorbidities (COPD, pneumonia, pulmonary emphysema, asthma, tuberculosis, coronary heart disease, hypertension, and stroke) were associated with a higher incidence of postoperative complications, as well as a higher risk of patient readmission to the healthcare facility after discharge.

Behinaein et al. [7] and Manzano et al. [8] have noted that complete surgical resection (lobar resection) remains the gold standard for the surgical treatment of NSCLC. In recent years, however, VATS procedures have taken a leading role in lung resection, especially in sub-lobar resections; as a less invasive approach, it was associated with fewer postoperative complications such as pain, a shorter hospital stay [9], and less lung deterioration postoperatively, thus improving quality of life through less resection of apparently healthy lung parenchyma [10]. Our study revealed that wedge lung resections were performed primarily by VATS (70.3% before PSM and 65.5% after PSM). We also found a higher frequency of postoperative complications after lobar resection surgery, with persistent pain being the most frequent complication (58% before PSM), as well as a longer hospital stay. Stamenovic et al. [3] found that patients undergoing atypical segmentectomy or pulmonary wedge resection had 3 times fewer postoperative complications than those who underwent lobectomy, and fewer hospital readmissions for the management of these complications.

A tumor size larger than 2 cm can lead to more significant complications in the postoperative period; however, lobectomy and sub-lobar pulmonary resection (wedge or segmentectomy) of tumors smaller than 2 cm can have similar long-term mortality results [11]. Our study found that the most common histological type of early-stage NSCLC was adenocarcinoma (66.65% before PSM). Similar results were reported by Altorki et al. [12] and Li et al. [4], who found that 63.7% and 73.6%, respectively, of lung cancers had this same histological class of tumors. Tumors measuring 1–2 cm were the most frequently found in our study during the diagnostic process (31.3% before PSM and 31.6% after PSM); Altorki et al. [12] found that a tumor size of 1–1.5 cm represented 50.8% of all lung cancers.

When comparing wedge resection and lobectomy for 5-year outcomes, we found no statistically significant differences regarding locoregional recurrence, even after PSM. In addition, both groups showed similar OS and DFS results, including in the combined analysis with the margin size (10 mm) variable. Though we found before PSM that the lobectomy group had a higher percentage (91.1%) of LRFS compared to the wedge resection group (79.8%), after PSM the difference was smaller and not statistically significant. However, the results regarding LRFS were significant after PSM, when the combined analysis was carried out considering subgroups by margin size and lymph node removal, both with favorable results for the wedge resection group (Table 1). The lobectomy group had a lower likelihood of LRFS when we considered variables such as margins and lymph node resection and adjusted for confounding factors; nevertheless, one reason may be the histological type and any micro-metastasis not identified in the pathological study [1,6,8]. Another possibility is histological differentiation, which may be a pivotal factor to consider in future studies [12].

In recent years, several studies and reviews around the world have been published that show similarities with our results [13]. For example, in 2022, Shi et al. [14] published a Bayesian meta-analysis comparing the OS, DFS, and relapse-free survival (RFS) outcomes of wedge resection and lobectomy/segmentectomy for treatment of early-stage NSCLC. Patients with early-stage NSCLC who received lobectomy had a lower OS hazard ratio than patients who received wedge resection, indicating that the OS was longer in patients who received lobectomy than in those who received wedge resection. However, the surgical approaches showed no significant difference in DFS and RFS. Song et al. [15] found that wedge resection was comparable to lobectomy regarding OS when treating stage IA second-primary NSCLC patients with previous lung cancer-directed surgery, but they suggested that wedge resection might be sufficient for such patients, although wedge resection could have a higher lung cancer-specific mortality rate than lobectomy in some cases.

Similar results were presented by Ding et al. [16], who found that wedge resection was inferior to lobectomy for NSCLC ≤1 cm and 1–2 cm; however, wedge resection plus adequate lymph node resection was comparable to lobectomy. In an Italian study with 538 participants, Bertolaccini et al. [17] showed that lobectomy and wedge resection were equivalent oncologic treatments for individuals with cN0/cM0 stage IA NSCLC <2 cm. Furthermore, Lv et al. [18] explained why sub-lobectomy (segmentectomy and wedge resection) had emerged as an alternative modality; in their systematic comparison of the treatment effects of lobectomy and sub-lobectomy regarding the survival outcomes of patients with stage I NSCLC, the patients experienced poor OS, whereas their DFS was similar for both methods.

A few systematic reviews have been published with important results. For example, Divisi et al. [19] found a higher rate of complications in patients who underwent lobectomy compared to sub-lobar resection, especially for cases with a high comorbidity index or octogenarian patients. The relapse rate and OS were more favorable for patients who underwent a lobectomy. They recommended sublobar resections for patients who were elderly, had a high comorbidity index, or had a low respiratory functional reserve [4,7]. However, pulmonary lobectomy is still considered the safest and oncologically correct method in patients with good performance status or a high risk of recurrence. In contrast, in a retrospective cohort study at the Lancashire Cardiac Centre, United Kingdom, Moon et al. [20] showed that non-anatomical wedge resection was associated with significantly less favorable 5-year survival compared to lobectomy in matched patients. They concluded that lobectomy should remain the standard of care for patients with early-stage lung cancer. A similar recommendation was offered by Zhang et al. [21] in their investigation with similar results (56.6% versus 61.32%, p=0.02).

The first report to compare conditional survival rates following lobectomy, segmentectomy, and wedge resection for patients with stage IA1 NSCLC was published in 2020 by Fan et al. [22]. Their findings indicated that lobectomy was the most conservative surgical procedure for stage IA1 NSCLC and raised questions regarding the improved long-term prognosis of segmentectomy in some subsets of patients [22]. Salazar et al. [23] found that the survival advantage of lobectomy over wedge resection for stage I NSCLC seemed to dissipate among patients who had a shorter life expectancy due to age and comorbidities. Because of this, they recommended wedge resection as a reasonable option for patients at high risk of dying from non-cancer-related causes.

In 2022, Song et al. [24] analyzed the prognostic significance of lung adenocarcinoma classification for patients with pathological N0 lung invasive adenocarcinomas ≤1 cm who underwent surgical resection. After investigating surgical procedures, they found that anatomical resection had a better survival rate for patients with a pN0 invasive lung adenocarcinoma ≤1 cm. Moreover, for patients with an invasive adenocarcinoma ≤1 cm for whom anatomical resection was not feasible, wedge resection could provide a similar oncological effect when the tumor was lepidic predominant or acinar predominant, and lymph node resection was considered effective when it involved between 3 and 4 stations [12,24]. Kamigaichi et al. [25] included 669 patients in their investigation of the radiological diagnostic factors for low-grade lung cancer and concluded that the radiologic criteria for ground-glass opacity and a low maximum standardized uptake value could predict low-grade cancer, even in solid-dominant NSCLC sized ≤2 cm. As a result, wedge resection could be an acceptable surgical option for patients with radiologically predicted indolent NSCLC showing a solid-dominant appearance, together with the possibility of effective lymph node resection of >5 nodes per surgical procedure [25].

A recent systematic review and meta-analysis published by Ling et al. [26] suggested that segmentectomy and lobectomy yielded similar oncological outcomes. However, wedge resection was associated with a poorer prognosis (<45% survival in 5 years, p=0.01). In contrast, for elderly patients, wedge resection could be a reasonable surgical option when accompanied by a wide lymph node resection (>5 nodes approximately) [26]. Similarly, Guo et al. [27] suggested that segmentectomy and lobectomy had comparable oncologic efficacy for solid-dominant stage IA cases of NSCLC (67.2% versus 71.9%, p=0.002), but segmentectomy may be a feasible alternative in selected cases [28]. Considering the roles of neoadjuvant and adjuvant therapies in early-stage NSCLC is important as well because therapies such as Durvalumab could have a crucial impact on long-term survival, with a crude odds ratio of 16.0 (95% confidence interval, 3.2–79.6; p<0.0001) for nonsurgical approaches such as stereotactic body radiotherapy [29].

In conclusion, though no difference was observed regarding OS or DFS when comparing wedge resection to lobectomy before our PSM analysis, margins of >10 mm were needed with the removal of ≥4 lymph nodes to confer LRFS outcomes similar to those following a lobectomy. When margins >10 mm were obtained and appropriate lymph node dissection was performed, patients undergoing wedge resection experienced fewer complications and could have similar survival outcomes to those undergoing lobectomy.

Author contributions

All authors contributed to the conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing–original draft, and writing–review & editing. All authors approved the final manuscript.

Conflict of interest

No potential conflict of interest relevant to this article was reported.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.