Prognostic impact of cancer-directed surgery in primary pulmonary lymphoma: a population-based, propensity score-matched analysis

Introduction

Primary pulmonary lymphoma (PPL) is a rare lymphoproliferative disorder characterized by clonal lymphoid infiltration restricted to the lung parenchyma and/or bronchi in the absence of extrathoracic involvement at diagnosis or within the subsequent 3 months (1,2). PPL accounts for fewer than 1% of all non-Hodgkin lymphomas (NHLs) and approximately 0.5–1% of primary pulmonary malignancies (1-3). Among the histologic subtypes, mucosa-associated lymphoid tissue (MALT) lymphoma is the most prevalent and typically exhibits an indolent clinical course, whereas diffuse large B-cell lymphoma (DLBCL) and T/NK-cell lymphomas manifest more aggressive behavior and poorer outcomes (2,4).

The rarity of PPL hampers the understanding of its optimal management. Most available evidence originates from single-center, small-sample retrospective studies, which restricts generalizability and impedes the establishment of standardized management strategies (5-8). Treatment options include watch-and-wait observation, surgical resection, chemotherapy, radiotherapy, immunotherapy, and multimodal approaches; however, no consensus has been reached regarding the optimal therapeutic strategy (2,8,9). In particular, the survival benefit of cancer-directed surgery (CDS) is debated, with some studies suggesting potential advantages in early-stage or localized disease, while others report no significant difference compared with nonsurgical management (7-11).

Large, population-based cancer registries such as the Surveillance, Epidemiology, and End Results (SEER) program offer an opportunity to overcome the inherent limitations of small institutional cohorts. SEER provides comprehensive patient-level data—including histologic subtype, stage, treatment codes, and long-term survival—allowing more robust epidemiologic and prognostic analyses for rare tumors such as PPL (12,13). Although SEER-based studies have begun to characterize prognostic factors and develop survival models for PPL, the impact of CDS has not been rigorously evaluated using analytical methods capable of addressing baseline confounding (13).

Accordingly, this study used SEER Research Plus data [2000–2018] to investigate the prognostic effect of CDS in patients with PPL. We applied propensity score matching (PSM) to reduce confounding by indication and employed an information-theoretic, AICc-based model-averaging framework to identify robust survival predictors. Our findings aim to generate reliable, population-level evidence regarding the potential survival benefits of CDS and to inform future risk-adapted treatment strategies for PPL. We present this article in accordance with the STROBE reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2539/rc).

Methods

Data collection

Data were obtained from the SEER Research Plus Database (18 registries; November 2020 submission, 2000–2018) using SEER*Stat software (version 8.4.0.1; https://seer.cancer.gov/data-software/). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Patients diagnosed with PPL between January 1, 2000, and December 31, 2014, were identified. Eligible cases required histologic confirmation and were retrieved using International Classification of Diseases for Oncology, 3rd Edition (ICD-O-3) histology codes: 9590–9599 [diffuse/not otherwise specified (NOS) malignant lymphomas]; 9650–9669 [Hodgkin lymphoma (HL)]; 9670–9699 (mature B-cell NHL); 9700–9719 (mature T-/NK-cell NHL); and 9720–9729 (precursor lymphoblastic NHL). Only tumors arising in the lung or bronchus, defined by primary site codes C34.0–C34.9, were included.

Clinical data extracted included patient identification number, age at diagnosis, sex, marital status, race/ethnicity, year of diagnosis, histologic subtype, primary site, laterality, presence of B symptoms, Ann Arbor stage, CDS, radiotherapy, chemotherapy, vital status, survival months, cause of death according to SEER, indication of initial malignancy, and confirmation of diagnosis.

In the SEER database, CDS is recorded as a binary variable indicating whether a surgical procedure directed at the primary tumor was performed. Detailed information regarding surgical intent (diagnostic versus definitive), extent of resection (e.g., wedge resection, lobectomy, pneumonectomy), margin status, and perioperative outcomes is not available. Accordingly, CDS in this study represents a heterogeneous category that may include both diagnostic and therapeutic procedures.

In the SEER database, radiotherapy is recorded as a binary variable indicating receipt of treatment but does not specify treatment timing, sequencing relative to surgery, or intent (adjuvant, definitive, or palliative). Accordingly, radiotherapy was modeled as a covariate rather than a time-dependent or sequence-specific exposure.

Data processing and variable definitions

Patients were excluded if they met any of the following criteria: (I) lack of microscopic confirmation; (II) diagnosis made only through autopsy or death certificate; (III) missing key clinical information (race, marital status, Ann Arbor stage); (IV) more than one primary malignancy; (V) survival time of 0 months or missing data; or (VI) unknown or missing cause of death precluding cancer-specific survival (CSS) assessment.

Histologic subtypes were categorized as DLBCL, follicular lymphoma (FL), MALT lymphoma, HL, lymphoplasmacytic lymphoma (LPL), chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL), mantle cell lymphoma (MCL), T/NK-cell NHL (grouped), and others (including primary effusion lymphoma, intravascular large B-cell lymphoma, Burkitt lymphoma, NHL-NOS, and malignant lymphoma-NOS).

The follow-up cutoff date was December 31, 2014, ensuring a minimum potential follow-up period of 5 years. Overall survival (OS) was defined as the interval from diagnosis to death from any cause. CSS was defined as the interval from diagnosis to death attributable to PPL, with deaths from other causes treated as censored events.

PSM

PSM was employed to minimize baseline confounding between the CDS and non-CDS groups. Propensity scores were estimated using multivariable logistic regression incorporating all demographic and clinical variables. A 1:1 nearest-neighbor matching algorithm without replacement was performed using a caliper width of 0.10 of the standard deviation of the logit of the propensity score. Covariate balance was assessed using standardized mean difference (SMD), with SMD values less than 0.10 indicating acceptable balance.

Statistical analysis

All analyses were conducted using R software (version 4.2.1). Continuous variables were compared using Student’s t-tests, and categorical variables were compared using χ² tests before and after PSM. Survival curves for OS and CSS were generated using the Kaplan-Meier method and compared using the log-rank test. Cox proportional hazards models were fitted to estimate hazard ratios (HRs) and 95% confidence intervals (CIs), with ties handled using the Breslow method.

To evaluate prognostic performance, 13 candidate Cox models were constructed, ranging from null to fully saturated models. The corrected Akaike information criterion (AICc) was used to identify the best-fitting model, and a 95% confidence model set was generated based on AICc weighting. Model-averaged HRs were calculated to identify robust survival predictors. A two-sided P value less than 0.05 was considered statistically significant.

Results

Patient characteristics

A total of 2,782 patients with PPL met the inclusion criteria (Figure 1). Baseline characteristics are summarized in Table 1. The median age was 64.5 years [standard deviation (SD) ±16.3 years], and 53.6% of patients were female. The majority of patients were White (86.0%). MALT lymphoma was the most common histologic subtype (35.8%), followed by DLBCL (33.7%). Most tumors were unilateral (86.8%), and Ann Arbor stage I was most prevalent (43.2%). B symptoms were present in 7.2% of patients. Radiotherapy was administered to 13.6% and chemotherapy to 57.8% of patients.

Figure 1 Patient selection flowchart. The diagram illustrates the systematic selection process from the initial SEER database query to the final propensity score-matched cohort. CDS, cancer-directed surgery; CSS, cancer-specific survival; PPL, primary pulmonary lymphoma; SEER, Surveillance, Epidemiology, and End Results.

A total of 889 patients (32.0%) underwent CDS, whereas 1,893 (68.0%) did not. Before PSM, significant imbalances were observed in age, sex, marital status, histologic subtype, tumor site, laterality, Ann Arbor stage, B symptoms, radiotherapy, and chemotherapy (all P<0.05). After 1:1 matching, 727 well-balanced pairs (n=1,454) were generated. As shown in Table 2, all covariates achieved acceptable balance (all P>0.05), indicating effective reduction of baseline confounding. Ann Arbor stage was explicitly included in the PSM, and post-matching balance was achieved for disease stage, with standardized mean differences below 0.10 and no significant differences between groups.

Survival outcomes after PSM

AICc-based model selection identified an optimal model comprising age, sex, CDS, histologic subtype, Ann Arbor stage, radiotherapy, and marital status (Table 3). Multivariable Cox regression confirmed these variables as independent prognostic factors (Figure 2).

Figure 2 Forest plot of multivariable Cox regression analysis for cancer-specific survival. HRs and 95% CIs are shown for each prognostic factor in the propensity score-matched cohort (n=1,454). Variables associated with inferior CSS are displayed above the reference line (HR >1), while protective factors are shown below (HR <1). *, P<0.05; **, P<0.01; ***, P<0.001. AIC, Akaike information criterion; CDS, cancer-directed surgery; CI, confidence intervals; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic lymphoma; CSS, cancer-specific survival; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; HL, Hodgkin lymphoma; HRs, hazard ratios; LPL, lymphoplasmacytic lymphoma; MALT, mucosa-associated lymphoid tissue lymphoma; MCL, mantle cell lymphoma; T/NK cell NHL, T/natural killer cell non-Hodgkin lymphoma.

For CSS, independent predictors of inferior survival included male sex (HR 1.30, 95% CI: 1.05–1.61), T/NK-cell NHL (HR 4.13, 95% CI: 2.04–8.36), Ann Arbor stage III–IV disease (HR 1.66, 95% CI: 1.34–2.05), and absence of CDS (HR 1.60, 95% CI: 1.30–1.97). Favorable predictors included age 60 years or younger (HR 0.37, 95% CI: 0.28–0.48), married status (HR 0.75, 95% CI: 0.58–0.97), MALT lymphoma (HR 0.38, 95% CI: 0.21–0.72), and receipt of radiotherapy (HR 0.56, 95% CI: 0.34–0.91).

The Cox model for OS demonstrated similar patterns (Figure 3). Male sex, T/NK-cell NHL, Ann Arbor stage III–IV, and absence of CDS were associated with inferior OS, whereas younger age, married status, MALT, FL, HL, and LPL subtypes, and radiotherapy were associated with superior OS.

Figure 3 Forest plot of multivariable Cox regression analysis for overall survival. HRs and 95% CIs are shown for each prognostic factor in the propensity score-matched cohort (n=1,454). Variables associated with inferior OS are displayed above the reference line (HR >1), while protective factors are shown below (HR <1). *, P<0.05; **, P<0.01; ***, P<0.001. AIC, Akaike information criterion; CDS, cancer-directed surgery; CI, confidence intervals; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic lymphoma; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; HL, Hodgkin lymphoma; HRs, hazard ratios; LPL, lymphoplasmacytic lymphoma; MALT, mucosa-associated lymphoid tissue lymphoma; MCL, mantle cell lymphoma; OS, overall survival; T/NK cell NHL, T/natural killer cell non-Hodgkin lymphoma.

Survival analyses in the matched cohort

We evaluated the prognostic impact of CDS by comparing CSS and OS between propensity score-matched groups (n=727 pairs). Patients who underwent CDS demonstrated significantly superior CSS compared with those who did not receive CDS (Figure 4A). Subgroup analyses revealed significant survival differences according to age, sex, marital status, Ann Arbor stage, and radiotherapy status (Figure 4B-4F). Specifically, CSS benefits associated with CDS were observed among patients aged 60 years or younger, females, married individuals, those with Ann Arbor stage I–II disease, and those who received radiotherapy. Consistent across subgroup analyses and multivariable models, the survival benefit associated with CDS was most evident among patients with early-stage (Ann Arbor I–II) disease.

Figure 4 Kaplan-Meier curves comparing cancer-specific survival between CDS and non-CDS groups in the matched cohort. (A) Overall matched cohort (n=727 pairs). Subgroup analyses stratified by (B) age (≤60 vs. >60 years), (C) sex (male vs. female), (D) marital status (married vs. unmarried vs. divorced), (E) Ann Arbor stage (stages I–II vs. stages III–IV), and (F) receipt of radiotherapy (yes vs. no). P values were calculated using the log-rank test. CDS, cancer-directed surgery; RT, radiotherapy.

Figure 5 demonstrates that patients with MALT lymphoma who underwent CDS experienced improved CSS. Subgroup analyses further demonstrated significant heterogeneity in the association between CDS and survival across histologic subtypes. A significant CSS benefit associated with CDS was observed primarily among patients with MALT lymphoma, whereas no consistent survival advantage was observed for high-grade histologies, including DLBCL and T/NK-cell lymphomas.

Figure 5 Kaplan-Meier curves comparing cancer-specific survival between CDS and non-CDS groups stratified by histologic subtype. (A) MALT lymphoma, (B) DLBCL, (C) T/NK-cell NHL, (D) FL, and (E) HL. P values were calculated using the log-rank test. CDS, cancer-directed surgery; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; HL, Hodgkin lymphoma; MALT, mucosa-associated lymphoid tissue; NHL, non-Hodgkin lymphoma.

Patients in the CDS group also demonstrated superior OS compared with those in the non-CDS group (Figure 6A). OS benefits associated with CDS were observed among patients aged 60 years or younger, females, married individuals, and those with Ann Arbor stage I–II disease (Figure 6B-6F). No significant differences in OS were observed across histologic subtypes in either the CDS or non-CDS groups (Figure 7).

Figure 6 Kaplan-Meier curves comparing overall survival between CDS and non-CDS groups in the matched cohort. (A) Overall matched cohort (n=727 pairs). Subgroup analyses stratified by (B) age (≤60 vs. >60 years); (C) sex (male vs. female); (D) marital status (married vs. unmarried vs. divorced), (E) Ann Arbor stage (stages I–II vs. stages III–IV), and (F) receipt of radiotherapy (yes vs. no). P values were calculated using the log-rank test. CDS, cancer-directed surgery; RT, radiotherapy.

Figure 7 Kaplan-Meier curves comparing overall survival between CDS and non-CDS groups stratified by histologic subtype. (A) MALT lymphoma, (B) DLBCL, (C) T/NK-cell NHL, (D) FL, and (E) HL. P values were calculated using the log-rank test. CDS, cancer-directed surgery; DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; HL, Hodgkin lymphoma; MALT, mucosa-associated lymphoid tissue; NHL, non-Hodgkin lymphoma.

Discussion

In this population-based, propensity score-matched analysis of 2,782 patients with PPL from the SEER Research Plus Database, CDS was associated with significantly improved CSS and OS after adjustment for measured confounders. This association remained robust in the matched cohort and in multivariable, AICc-guided, model-averaged Cox regression analyses. Subgroup analyses demonstrated that survival benefits associated with CDS were most pronounced among patients aged 60 years or younger, females, married individuals, those with early-stage disease (Ann Arbor stages I–II), and those who received radiotherapy. Among histologic subtypes, patients with MALT lymphoma derived a particular CSS benefit from CDS. These findings contribute substantial population-level evidence to a literature predominantly composed of small institutional series with conflicting conclusions regarding the role of surgery in PPL (1,9,14-16).

Notably, before PSM, patients who underwent surgery were substantially more likely to present with Ann Arbor stage I disease, reflecting clinical selection of patients with localized disease and raising the possibility of confounding by indication.

PPL comprises a biologically heterogeneous spectrum of diseases with distinct clinical behavior and therapeutic paradigms. Our findings highlight substantial heterogeneity in the association between CDS and survival across histologic subtypes. The observed survival benefit of surgery was most pronounced in patients with MALT lymphoma, an indolent subtype that often presents as localized disease amenable to complete resection. In contrast, for aggressive histologies such as DLBCL and T/NK-cell lymphomas, which are typically managed with systemic therapy, surgery did not confer a convincing survival advantage. These results suggest that the overall association between CDS and improved survival in the pooled PPL cohort is likely driven by low-grade, localized subtypes rather than high-grade disease.

Our findings align with clinical expectations and several prior reports demonstrating favorable outcomes in patients with localized, indolent PPL who undergo resection (4,8,17,18). MALT lymphoma, the predominant PPL subtype in our cohort, is typically indolent and frequently presents as a localized lesion amenable to surgical resection. Resection may achieve excellent local control and durable remission in selected patients (3,4,19). In contrast, high-grade histologies such as DLBCL and T/NK-cell lymphomas were associated with markedly inferior prognosis in our analysis, consistent with prior observational studies (20-25). The observed treatment-effect heterogeneity according to histologic subtype and Ann Arbor stage supports a tailored approach—surgery may offer greater benefit for localized, low-grade tumors such as MALT lymphoma compared with disseminated or biologically aggressive subtypes.

Despite these consistencies, prior reports have yielded conflicting conclusions regarding the benefit of surgery (9,26-28). Several single-center retrospective studies and small cohorts found no clear survival advantage for surgical resection in early-stage MALT lymphoma or reported equivalent outcomes with nonoperative approaches such as radiotherapy or immunochemotherapy (27-29). Differences between those studies and the current analysis likely reflect limited sample sizes, single-institution selection bias, and incomplete adjustment for confounding. By leveraging a large population-based registry and applying PSM combined with an information-theoretic model-averaging strategy, our study substantially reduces the influence of measured confounders and provides more generalizable estimates of the association between CDS and survival, although residual confounding cannot be eliminated.

Several non-mutually exclusive mechanisms may account for the observed association between CDS and improved survival. For localized, indolent PPL such as MALT lymphoma, complete surgical excision may eliminate the disease and thereby reduce the risk of disease-related complications and progression. Surgery may also facilitate definitive histopathologic diagnosis and more accurate staging, enabling appropriate selection of adjuvant therapy (29,30). Additionally, selection bias may contribute to observed outcomes: patients offered surgery tend to be younger, have fewer comorbidities, superior pulmonary reserve, and more localized disease. Although PSM mitigates measured imbalances, residual confounding by unmeasured factors such as performance status, pulmonary function, comorbidity burden, tumor size, and institutional practice patterns likely persists.

The observed interaction between CDS and radiotherapy likely reflects selection of patients with favorable disease characteristics for multimodality treatment rather than a causal effect of adjuvant radiotherapy following surgery. Because SEER does not provide information on radiotherapy timing or intent, we cannot distinguish postoperative adjuvant radiotherapy from definitive radiotherapy administered in nonsurgical patients. Consequently, the apparent survival benefit in patients receiving both CDS and radiotherapy should be interpreted cautiously and viewed as an association with combined treatment strategies in selected, often early-stage cases, rather than evidence supporting a specific therapeutic sequence.

Accordingly, the role of surgery in PPL should not be viewed as uniform across all subtypes but rather considered in a subtype-specific and stage-adapted context, with multidisciplinary evaluation guiding individualized treatment decisions.

Our results suggest that CDS should be considered a treatment option for selected patients with PPL, particularly those with localized, resectable MALT lymphoma who are medically fit for surgery. Decision-making should be multidisciplinary, incorporating histologic subtype, disease extent, pulmonary function, comorbidities, patient preferences, and availability of effective nonoperative alternatives such as radiotherapy or immunochemotherapy (31,32). Importantly, these findings do not suggest that all patients with PPL should undergo surgery but rather provide population-level evidence to inform individualized, risk-stratified treatment planning and guide the design of prospective trials.

This study has several strengths. First, the large sample size derived from a national population-based registry provides increased statistical power and enhanced generalizability compared with previous single-center series. Second, PSM reduced measured baseline differences between surgical and nonsurgical groups, with standardized mean differences after matching demonstrating acceptable covariate balance. Third, application of an AICc-based model-averaging framework reduced reliance on any single model specification and yielded robust estimates of prognostic factors across candidate models. Finally, both CSS and OS were evaluated, and subgroup analyses were conducted to explore heterogeneity of treatment effects according to age, sex, Ann Arbor stage, histologic subtype, and receipt of radiotherapy.

Several important limitations must be considered when interpreting these findings. First, the retrospective observational design precludes causal inference. Although PSM and multivariable adjustment were applied to reduce confounding by measured variables, residual confounding from unmeasured factors cannot be excluded. These include comorbidity burden, pulmonary function, performance status, tumor size and radiographic extent, symptom severity influencing treatment selection, and details of systemic therapy, such as chemotherapy regimens and use of immunotherapy (e.g., rituximab), which are not fully captured in the SEER database. Importantly, residual confounding related to disease stage—such as tumor burden and radiographic extent not fully reflected by Ann Arbor classification—may persist despite successful matching. Such unmeasured factors may influence both the likelihood of undergoing surgery and survival outcomes; therefore, the observed associations should be interpreted as associative rather than causal.

Second, the SEER database does not provide granular information regarding surgical modality or treatment intent. CDS is recorded as a binary variable without distinction between diagnostic, staging, palliative, or definitive procedures, nor does it capture the extent of resection, surgical margin status, or perioperative morbidity. Consequently, the CDS category represents a heterogeneous group that may include diagnostic surgical procedures, such as biopsies, which are unlikely to confer a therapeutic survival benefit. This potential misclassification may attenuate observed survival differences and complicate interpretation of the surgical effect. In addition, lack of information on radiotherapy sequencing and intent precludes evaluation of whether radiotherapy was administered as adjuvant therapy after resection or as definitive treatment in nonsurgical patients, limiting causal interpretation of combined-modality effects. Selection bias remains a concern in observational surgical studies. Patients selected for surgery may inherently represent a more favorable-risk population, characterized by younger age, better baseline health, superior pulmonary reserve, and more localized disease—factors not fully captured in SEER. Although covariate balance was substantially improved after matching, such bias cannot be entirely eliminated. Third, coding misclassification and missing data inherent in registry studies represent potential sources of error. Fourth, although the cohort was restricted to cases diagnosed through 2014 to ensure adequate follow-up, therapeutic paradigms have continued to evolve, including wider adoption of targeted agents and immunotherapies, which may limit applicability to patients treated under current standards of care. Fifth, some subgroup analyses, particularly those involving rare histologic subtypes, are based on small sample sizes and yield wide confidence intervals; these estimates should be interpreted with caution.

Importantly, these findings do not imply that all patients with PPL should undergo surgery. Rather, they suggest that surgery may be associated with improved outcomes in carefully selected patients, particularly those with localized, resectable disease. Multidisciplinary evaluation remains essential when considering surgical intervention. Prospective multicenter registries that capture granular clinical, radiologic, procedural, and molecular data are needed to better identify which patients with PPL benefit most from surgery. Where feasible, randomized or pragmatic trials comparing resection with organ-preserving strategies (radiotherapy, immunochemotherapy, or active surveillance) in well-defined cohorts with early-stage MALT lymphoma would provide the highest level of evidence (33,34). In the interim, pragmatic comparative effectiveness studies using detailed clinical datasets such as electronic health records or claims data linked to tumor registries, combined with advanced analytic techniques including instrumental variable analysis, inverse probability weighting, and competing-risk regression models, may help establish causal estimates. Development and external validation of prognostic nomograms integrating clinical, histologic, and treatment variables could guide individualized decision-making and inform trial design (13).

Conclusions

In this large population-based, propensity score-matched analysis, CDS was associated with improved CSS and OS in patients with PPL, particularly among younger, married patients with early-stage disease and those with MALT lymphoma. Although these findings support consideration of surgery for selected patients with resectable PPL, they should be interpreted cautiously given the nonrandomized study design and potential for residual confounding. Prospective studies and more comprehensive clinical registries are needed to validate these observations and inform evidence-based, risk-adapted management strategies for this rare malignancy.

Acknowledgments

The authors acknowledge the contributions of the National Cancer Institute and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER database. The interpretation and reporting of these data are the sole responsibility of the authors.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2539/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2539/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2539/coif). The 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. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

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. Cadranel J, Wislez M, Antoine M. Primary pulmonary lymphoma. Eur Respir J 2002;20:750-62. [Crossref] [PubMed]
2. Sanguedolce F, Zanelli M, Zizzo M, et al. Primary Pulmonary B-Cell Lymphoma: A Review and Update. Cancers (Basel) 2021;13:415. [Crossref] [PubMed]
3. Borie R, Wislez M, Thabut G, et al. Clinical characteristics and prognostic factors of pulmonary MALT lymphoma. Eur Respir J 2009;34:1408-16. [Crossref] [PubMed]
4. Wang L, Ye G, Liu Z, et al. Clinical characteristics, diagnosis, treatment, and prognostic factors of pulmonary mucosa-associated lymphoid tissue-derived lymphoma. Cancer Med 2019;8:7660-8. [Crossref] [PubMed]
5. Qian J, Luo DL, Zhang JE, et al. Diagnostic and prognostic factors for patients with primary pulmonary non-Hodgkin’s lymphoma: A 16-year single-center retrospective study. Oncol Lett 2019;18:2082-90. [Crossref] [PubMed]
6. Hu M, Gu W, Chen S, et al. Clinical Analysis of 50 Cases of Primary Pulmonary Lymphoma: A Retrospective Study and Literature Review. Technol Cancer Res Treat 2022;21:15330338221075529. [Crossref] [PubMed]
7. Husnain M, Kuker R, Reis IM, et al. Clinical and radiological characteristics of patients with pulmonary marginal zone lymphoma: A single center analysis. Cancer Med 2020;9:5051-64. [Crossref] [PubMed]
8. Xu Y, Zheng M, Guo Q, et al. Clinical Features and Survival Outcome of Early-Stage Primary Pulmonary MALT Lymphoma After Surgical Treatment. Front Surg 2021;8:713748. [Crossref] [PubMed]
9. Lin H, Zhou K, Peng Z, et al. Surgery and chemotherapy cannot improve the survival of patients with early-stage mucosa-associated lymphoid tissue derived primary pulmonary lymphoma. Front Oncol 2022;12:965727. [Crossref] [PubMed]
10. Cao J, Lin H, Peng Z, et al. Surgical management of stage IE/IIE primary pulmonary lymphomas: a propensity score matching study. J Thorac Dis 2025;17:134-49. [Crossref] [PubMed]
11. Ning Y, He H, Li Q, et al. The prognosis of patients with primary pulmonary mucosa-associated lymphoid tissue lymphoma: Treated with surgery or chemotherapy? Eur J Cardiothorac Surg 2024;65:ezae064. [Crossref] [PubMed]
12. Surveillance, Epidemiology, and End Results (SEER) Program. National Cancer Institute. Available online: www.seer.cancer.gov
13. Low SK, Zayan AH, Istanbuly O, et al. Prognostic factors and nomogram for survival prediction in patients with primary pulmonary lymphoma: a SEER population-based study. Leuk Lymphoma 2019;60:3406-16. [Crossref] [PubMed]
14. Graham BB, Mathisen DJ, Mark EJ, et al. Primary pulmonary lymphoma. Ann Thorac Surg 2005;80:1248-53. [Crossref] [PubMed]
15. Ferraro P, Trastek VF, Adlakha H, et al. Primary non-Hodgkin’s lymphoma of the lung. Ann Thorac Surg 2000;69:993-7. [Crossref] [PubMed]
16. Giza A, Dumnicka P, Kalicińska E, et al. Retrospective analysis of primary pulmonary lymphoma cases from Polish Lymphoma Research Group centers reveals associations between initial symptoms and outcomes. Sci Rep 2025;15:27883. [Crossref] [PubMed]
17. Vanden Eynden F, Fadel E, de Perrot M, et al. Role of surgery in the treatment of primary pulmonary B-cell lymphoma. Ann Thorac Surg 2007;83:236-40. [Crossref] [PubMed]
18. Shen H, Zhou Y. Clinical Features and Surgical Treatment of Primary Pulmonary Lymphoma: A Retrospective Study. Front Oncol 2022;12:779395. [Crossref] [PubMed]
19. Kim JH, Lee SH, Park J, et al. Primary pulmonary non-Hodgkin’s lymphoma. Jpn J Clin Oncol 2004;34:510-4. [Crossref] [PubMed]
20. He X, Huang Q, Li W, et al. Prognostic factors and predictive models for primary pulmonary diffuse large B-cell lymphoma: a population-based analysis. Hematology 2024;29:2420160. [Crossref] [PubMed]
21. Liu B, Liu H, Guo L, et al. Primary Pulmonary Diffuse Large B Cell Lymphoma Mimicking Metastasis: A Case Report and Literature Review. Onco Targets Ther 2020;13:5837-43. [Crossref] [PubMed]
22. Wang Y, Wang Z, Wu C, et al. Primary pulmonary extranodal NK/T-cell lymphoma: a case report and literature review. Transl Cancer Res 2020;9:7359-65. [Crossref] [PubMed]
23. Li Q, Zuo H, Liu C, et al. Survival and prognostic factors for primary lung extranodal NK/T-cell lymphoma: a retrospective study of data from China and the SEER database. Front Oncol 2025;15:1496735. [Crossref] [PubMed]
24. Li S, Huang Y, Wang L, et al. Primary pulmonary diffuse large B cell lymphoma presenting with features of organizing pneumonia: a case report. J Med Case Rep 2025;19:42. [Crossref] [PubMed]
25. Lu X, Zhou D, Zou L, et al. Pulmonary diffuse large B-cell lymphoma with concurrent organizing pneumonia: a case report. Front Med (Lausanne) 2025;12:1599268. [Crossref] [PubMed]
26. Oh SY, Kim WS, Kim JS, et al. Pulmonary marginal zone B-cell lymphoma of MALT type--what is a prognostic factor and which is the optimal treatment, operation, or chemotherapy?: Consortium for Improving Survival of Lymphoma (CISL) study. Ann Hematol 2010;89:563-8. [Crossref] [PubMed]
27. Kocatürk Cİ, Seyhan EC, Günlüoğlu MZ, et al. Primary pulmonary non-Hodgkin’s lymphoma: ten cases with a review of the literature. Tuberk Toraks 2012;60:246-53. [Crossref] [PubMed]
28. Yan W, Wu B, Liao AJ, et al. Watch-and-wait or immediate immunotherapy/immunochemotherapy in patients with phase IE primary pulmonary MALT lymphoma? A multicenter retrospective study. Ann Hematol 2021;100:709-14. [Crossref] [PubMed]
29. Wang B, Zhang C, Wang B, et al. Comparisons of Surgery and/or Chemotherapy in the Treatment of Primary Pulmonary Mucosa-Associated Lymphoid Tissue Lymphoma. Ann Thorac Cardiovasc Surg 2015;21:109-13. [Crossref] [PubMed]
30. Du C, Zhang J, Wei Y, et al. Retrospective Analysis of 9 Cases of Primary Pulmonary Mucosa-Associated Lymphoid Tissue Lymphoma and Literature Review. Med Sci Monit Basic Res 2018;24:233-40. [Crossref] [PubMed]
31. Cardenas-Garcia J, Talwar A, Shah R, et al. Update in primary pulmonary lymphomas. Curr Opin Pulm Med 2015;21:333-7. [Crossref] [PubMed]
32. Poletti V, Ravaglia C, Tomassetti S, et al. Lymphoproliferative lung disorders: clinicopathological aspects. Eur Respir Rev 2013;22:427-36. [Crossref] [PubMed]
33. Borie R, Wislez M, Antoine M, et al. Lymphoproliferative Disorders of the Lung. Respiration 2017;94:157-75. [Crossref] [PubMed]
34. Borie R, Wislez M, Antoine M, et al. Pulmonary mucosa-associated lymphoid tissue lymphoma revisited. Eur Respir J 2016;47:1244-60. [Crossref] [PubMed]