Histologic Grade 3 in Stage 1A Lung Adenocarcinoma: Survival Risks Comparable to Stage 1B

Abstract

Introduction

Early stage NSCLC presents a significant risk of recurrence despite surgical resection. The International Association for the Study of Lung Cancer–proposed histologic grading system identifies high-grade tumors (≥20% solid, micropapillary, or complex glandular patterns) as a potential prognostic tool. This study evaluates the prognostic significance of histologic grade 3 within stage IA non-mucinous adenocarcinoma and its comparability to stage IB.

Methods

We retrospectively analyzed 729 patients with stage I non-mucinous adenocarcinoma who underwent surgical resection (2010–2017). Tumors were graded per the International Association for the Study of Lung Cancer system. Kaplan-Meier survival curves and Cox proportional hazards models assessed overall survival (OS) and disease-free survival (DFS).

Results

Stage IA grade 3 tumors demonstrated significantly worse OS and DFS compared with grade 1 and 2 tumors (p < 0.001). Survival outcomes for stage IA grade 3 tumors were comparable to stage IB (p = 0.677 for OS, p = 0.248 for DFS). Stage IA grade 3 tumors also demonstrated a trend toward worse survival than stage IB grades 1 and 2. Grade 3 tumors exhibited sufficient risk stratification power, comparable to the already well-established risk factor of pleural invasion. Tumors with any high-grade patterns (>0%) had poorer survival than those without, but risks were less pronounced than the 20% cutoff.

Conclusions

Histologic grade 3 in stage IA identifies high-risk tumors comparable to stage IB, highlighting its potential for refining staging criteria and guiding adjuvant therapy. Validation studies are needed to confirm these findings.

Introduction

Lung cancer remains a leading cause of cancer-related mortality worldwide,¹ with a substantial number of recurrences even in early stage patients after surgical resection.2, 3, 4 Over the years, significant advancements have been made in the risk assessment of pulmonary adenocarcinoma. The 2015 WHO classification proposed a histologic pattern-based classification system that categorized adenocarcinomas into the following five major patterns: lepidic, acinar, papillary, solid, and micropapillary.⁵ This classification system correlated histologic patterns with prognosis, providing a basis for stratifying tumors into low-, intermediate-, and high-grade categories based on the predominant pattern.⁵

In addition to these histologic patterns, the American Joint Committee on Cancer (AJCC) Cancer Staging Manual eighth edition (2017) introduced a paradigm shift by excluding lepidic components from tumor size measurement for staging purposes in invasive non-mucinous adenocarcinoma.⁶ This change allowed staging to better reflect prognosis by focusing on non-lepidic, invasive portions of the tumor. Building on this, the 2021 WHO classification further refined the histologic assessment by formally recognizing the complex glandular pattern as part of high-grade patterns.⁷ Concurrently, the International Association for the Study of Lung Cancer (IASLC) pathology committee proposed a histologic grading system that considers both the predominant pattern and the proportion of high-grade patterns (solid, micropapillary, and complex glandular).⁸ This system demonstrated significant prognostic value in predicting overall survival (OS) and disease-free survival (DFS)⁸ and has since been validated in clinical practice.

Despite these advancements, a considerable proportion of patients with early stage lung adenocarcinoma experience recurrence after surgical treatment. This underscores the need for further stratification of high-risk patients within early stage disease. Pleural invasion, as one of the key pathologic risk factors, has long been incorporated into lung cancer staging, elevating tumors from T1 to T2 classification.⁶ Although the histologic grading system was only recently proposed in 2020⁸ and has been used in practice for a relatively short time, it is essential to evaluate the prognostic implications of grade 3 tumors, characterized by more than or equal to 20% high-grade patterns. Similar to pleural invasion, histologic grade 3 may potentially refine risk stratification within stage I.

This study aims to evaluate the prognostic impact of histologic grade 3 in stage IA non-mucinous adenocarcinoma and its relationship to current staging criteria, with the goal of improving the stratification of high-risk patients for tailored management and treatment strategies.

Material and Methods

Patient Data and Study Design

From January 2010 to December 2017, a total of 1295 patients with surgically resected NSCLC at Seoul St. Mary’s Hospital were retrospectively reviewed. Patients who received neoadjuvant therapy or presented with recurrent tumors were excluded, leaving 1181 cases of newly diagnosed and upfront resected NSCLC. Among these, 824 cases were classified as stage I according to the eighth edition of the AJCC staging manual. After excluding cases other than invasive non-mucinous adenocarcinoma, a total of 729 patients were included in the study (Fig. 1).

Figure 1.

Flowchart of case selection.

The standard surgical procedure for stage I lung cancer was anatomic resection, such as lobectomy or bilobectomy, with systematic dissection of three or more mediastinal lymph node stations. Sublobar resection was performed for high-risk patients with comorbidities or for ground-glass opacity nodules near the visceral pleura, with adequate resection margins exceeding the tumor diameter.

This study was approved by the institutional review board of Seoul St. Mary’s Hospital at the Catholic University of Korea (referral number: KC19RNSI0103), and the requirement for individual informed consent was waived due to the retrospective nature of the study.

Histologic Evaluation

All tumors were adequately sectioned for diagnosis through gross examination, with all tumors less than or equal to 3 cm entirely submitted for evaluation. For tumors more than 3 cm, at least five representative sections were reviewed. All hematoxylin and eosin-stained slides were examined for pattern classification by three pathologists (YES, MK, and KYL) who were blinded to the patients’ clinical outcomes. In cases of disagreement, grading and pattern classification were finalized through consensus. To enhance consistency and precision in grading, whole slide images were generated using the Philips IntelliSite Ultra Fast Scanner (Philips Healthcare, Best, Netherlands) at ×40 magnification. Tumor regions were manually annotated using digital pathology software, allowing precise measurement of each histologic pattern in square millimeters (mm²) (Supplementary Fig. 1). The pathologists jointly evaluated these annotated images to determine the proportion of each histologic pattern, ensuring an objective and standardized grading process. This approach minimized interobserver variability and improved the reproducibility of histologic classification.

The histologic analysis and grading of invasive non-mucinous adenocarcinomas were performed according to the proposal by the IASLC Pathology Committee.⁸ Tumors were classified as follows: grade 1 for lepidic predominant, grade 2 for acinar or papillary predominant, and grade 3 for tumors with more than or equal to 20% high-grade patterns, including solid, micropapillary, or complex glandular patterns. Pathologic risk factors such as lymphovascular invasion and pleural invasion were also evaluated. Elastic special staining and podoplanin immunohistochemistry were performed when necessary to confirm the presence of these risk factors.

Statistical Analysis

Statistical analyses were performed using the latest version of R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria). OS and DFS were estimated using Kaplan-Meier survival curves, and differences between groups were compared using the log-rank test. Cox proportional hazards regression analysis was used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) to evaluate the impact of histologic grade and other risk factors on survival outcomes. p value less than 0.05 was considered significant.

Results

Patient Characteristics

This study included 729 patients diagnosed with stage I non-mucinous adenocarcinoma who underwent surgical resection between January 2010 and December 2017. Their clinical characteristics are summarized in Table 1. The mean age of the patients was 63 years (±10.2), and the cohort had a slight predominance of females (60.8%). Most patients were never smokers (71.3%), whereas 28.7% were former or current smokers. Among the smokers, the median pack-year (py) was 30 (Q1–Q3: 20–40) py.

Table 1.

Clinicopathologic Characteristics of Patients With Stage I Non-Mucinous Adenocarcinoma

Characteristics	N = 729
Age	63.2 (±10.2)
Sex
Male	286 (39.2%)
Female	443 (60.8%)
Smoking
Never smoker	520 (71.3%)
Former or current smoker	209 (28.7%)
Pack-year (smokers only), median (Q1–Q3)	30 (20–40) py
Operation
Wedge resection	106 (14.5%)
Segmentectomy	75 (10.3%)
Lobectomy	537 (73.7%)
Bilobectomy	11 (1.5%)
Stage (eighth AJCC)
IA1	276 (37.9%)
IA2	222 (30.5%)
IA3	67 (9.2%)
IB	164 (22.5%)
Histologic grade
Grade 1	332 (45.5%)
Grade 2	204 (28%)
Grade 3	192 (26.3%)
Pleural invasion
Intraparenchymal or PL0	602 (82.6%)
≥ PL1	127 (17.4%)
Lymphovascular invasion
Absent	525 (72.0%)
Present	204 (28.0%)

AJCC, American Joint Committee on Cancer.

Regarding surgical methods, lobectomy was the most common procedure, performed in 73.7% of patients, followed by wedge resection (14.5%), segmentectomy (10.3%), and bilobectomy (1.5%). On the basis of the eighth AJCC staging system, the distribution of stage IA subcategories was as follows: IA1 (37.9%), IA2 (30.5%), and IA3 (9.2%), with the remaining 22.5% classified as stage IB. Histologic grading revealed that 45.5% of tumors were grade 1, 28% were grade 2, and 26.3% were grade 3. Pleural invasion (≥PL1) was observed in 17.4% of the cases, whereas lymphovascular invasion was present in 28% of the tumors.

Pattern Distribution and Histologic Grade Distribution by Stage

On the basis of the proposal from the IASLC Pathology Committee,⁸ the proportion of high-grade patterns, including micropapillary, solid, and complex glandular patterns (Fig. 2A–D, Supplementary Fig. 1), was assessed. By consensus among pathologists, tumors with high-grade patterns accounting for more than 20% of the total tumor were classified as grade 3.

Figure 2.

Histologic examples of non-mucinous adenocarcinoma containing high-grade patterns. (A) Low-power view of a tumor with >20% high-grade patterns (outlined in red); (B) micropapillary, (C) solid, and (D) complex glandular patterns shown at high power.

The distribution of specific histologic patterns across different grades was analyzed (mean ± SD, median [Q1–Q3]%). In grade 1 tumors, the proportion of lepidic pattern was 76.0 ± 17.2% (80 [60–90]%), whereas in grade 2 tumors, the proportion of acinar/papillary pattern was 76.4 ± 13.1% (75 [70–90]%). In grade 3 tumors, high-grade patterns (solid, micropapillary, complex glandular) accounted for 47.7 ± 26.3% (40 [25–70]%) (Supplementary Table 1).

In stage IA (N = 565), most patients (81.2%, N = 459) had grade 1 or 2 tumors, whereas 18.8% (N = 106) had grade 3 tumors. In stage IB (N = 164), grade 1 and 2 tumors were found in 47.6% (N = 78), whereas 52.4% (N = 86) had grade 3 tumors. When considering tumors with any high-grade patterns, 370 stage IA cases (65.5%) did not include any high-grade patterns, whereas 195 cases (34.5%) included at least some high-grade patterns. Similarly, in stage IB, 44 cases (26.8%) were classified without high-grade patterns, whereas 120 cases (73.2%) included high-grade patterns.

Analysis of Histologic Grade 3 and Other Risk Factors

Cox proportional hazards regression analysis was conducted to evaluate the impact of histologic grade 3 (≥20% high-grade patterns), the presence of any high-grade patterns, and other risk factors on survival outcomes in stage I non-mucinous adenocarcinoma. The detailed results are summarized in Table 2.

Table 2.

Comparison of Histologic Grade 3 in Stage 1A Lung Adenocarcinoma with Other Risk Factors

Patient Subgroup	Overall Survival		Disease-Free Survival
	Hazard Ratio	p Value	Hazard Ratio	p Value
Stage IA grades 1 and 2
Stage IA grade 3	14.03 (3.799–51.85)	<0.001	8.65 (4.47–16.72)	<0.001
Stage IB	16.58 (4.792–57.40)	<0.001	11.58 (6.36–21.07)	<0.001
Stage IA without high-grade pattern
Stage IA with any high-grade pattern	23.31 (3.009–180.6)	0.003	40.37 (9.718–167.7)	<0.001
Stage IB	41.07 (5.417–311.3)	<0.001	67.28 (16.327–277.2)	<0.001
≤3 cm without pleural invasion (stage IA)
≤3 cm with pleural invasion (stage IB)	4.84 (2.00–11.70)	<0.001	4.52 (2.75–7.42)	<0.001
>3 cm (stage IB)	8.73 (3.21–23.72)	<0.001	5.97 (3.17–11.24)	<0.001
Stage IA without lymphovascular invasion
Stage IA with lymphovascular invasion	12.78 (3.450–47.30)	<0.001	5.85 (3.070–11.14)	<0.001
Stage IB	16.45 (4.744–57.01)	<0.001	9.83 (5.561–17.37)	<0.001

In stage IA, grade 3 tumors (≥20% high-grade patterns) had a significantly higher risk of worse OS and DFS compared with grade 1 and 2 tumors (<20% high-grade patterns). Similarly, tumors with any high-grade patterns (>0%) demonstrated markedly worse outcomes compared with those without high-grade patterns. To better understand the prognostic significance of histologic grade 3, we compared its impact on survival to other established risk factors, including pleural invasion and lymphovascular invasion.

Survival Implications of Stage IA Grade 3 in Comparison to Stage IB in Invasive Non-Mucinous Adenocarcinoma

We assessed the survival outcomes of stage IA grade 3 tumors by comparing them with stage IA grade 1 and 2 and stage IB tumors (Fig. 3A and B). A significant difference was observed between stage IA grade 1 and 2 and stage IA grade 3 (p < 0.001). However, no significant difference in risk was observed between stage IA grade 3 and stage IB (p = 0.677 for OS, p = 0.248 for DFS). Interestingly, for OS, stage IA grade 3 tumors demonstrated a trend toward worse survival compared with stage IB, although the difference was not statistically significant (Fig. 3A). Furthermore, when stage IB was stratified into grade 1 and 2 and grade 3 subgroups, stage IA grade 3 tumors demonstrated a trend toward worse survival outcomes compared with stage IB grades 1 and 2, a finding consistent across both OS and DFS (Fig. 3C and D). In addition, stage IA grade 3 and stage IB grade 3 tumors had no significant difference in OS (p = 0.102) but a significant difference in DFS (p = 0.013), suggesting that the IA versus IB staging distinction remains prognostically relevant, particularly for DFS, even within grade 3 tumors.

Figure 3.

The impact of histologic grade 3 on survival in stage I lung adenocarcinoma compared with the influence of pleural invasion. (A, B) Kaplan-Meier survival curves comparing stage IA grades 1 and 2, stage IA grade 3, and stage IB for OS (A) and DFS (B). (C, D) Kaplan-Meier survival curves comparing stage IA grades 1 and 2 (1AG12), stage IA grade 3 (1AG3), stage IB grades 1 and 2 (1BG12), and stage IB grade 3 (1BG3) for OS (C) and DFS (D). Log-rank test p values: 1AG3 versus 1BG12: p = 0.142 (OS), p = 0.485 (DFS); 1BG12 versus 1BG3: p = 0.005 (OS), p = 0.005 (DFS); 1AG3 versus 1BG3: p = 0.102 (OS), p = 0.013 (DFS). (E, F) Kaplan-Meier survival curves comparing tumors less than 3 cm without pleural invasion (stage IA), tumors less than 3 cm with pleural invasion, and tumors more than or equal to 3 cm for OS (E) and DFS (F). (G, H) Kaplan-Meier survival curves comparing tumors less than 3 cm without pleural invasion (stage IA), tumors less than 3 cm with pleural invasion, tumors more than or equal to 3 cm without pleural invasion, and tumors more than or equal to 3 cm with pleural invasion for OS (G) and DFS (H). DFS, disease-free survival; OS, overall survival.

To compare the prognostic impact of histologic grade 3 with the well-established risk factor pleural invasion, we evaluated its effect on survival outcomes. As previously known, tumors less than or equal to 3 cm with pleural invasion had significantly worse survival compared with those without pleural invasion.⁹ Furthermore, no significant survival difference was observed between tumors less than or equal to 3 cm with pleural invasion and tumors more than 3 cm (Fig. 3E and F), a pattern similar to the risk contribution of histologic grade 3. However, unlike grade 3, which stratified risk within stage IB, pleural invasion failed to stratify risk further within stage IB (Fig. 3G and H).

For tumors with any high-grade patterns (>0%), regardless of the 20% cutoff used for grade 3 classification, significantly worse OS and DFS were observed compared with those without any high-grade patterns in stage IA (Supplementary Fig. 2). In comparison with stage IB, no significant difference in OS was observed between stage IA with any high-grade patterns and stage IB (p = 0.148). However, for DFS, a significant survival difference was maintained, with stage IA tumors having better outcomes than stage IB (p = 0.020).

Discussion

Identifying high-risk patients within early stage NSCLC who are eligible for curative surgical resection is crucial. Recent evidence has demonstrated significant OS benefits with adjuvant osimertinib in EGFR-mutated, completely resected NSCLC,¹⁰ including the efficacy of adjuvant immunotherapy in early stage NSCLC.^11,12 These benefits are primarily observed in patients classified as stage IB to IIIA.10, 11, 12 Consequently, the distinction between stage IA and IB may hold clinical significance in stratifying patients for tailored adjuvant treatment strategies.

The pathologic staging of lung cancer has undergone continuous evolution and refinement. Since the publication of the first AJCC Staging Manual in 1977,¹³ pleural invasion has been recognized as a critical factor, elevating tumors to T2 classification even when tumor size is less than 3 cm. This criterion has been consistently maintained, reflecting the substantial increase in risk associated with accurately defined visceral pleural invasion.⁹ Furthermore, until the introduction of the eighth edition of the AJCC Staging Manual in 2017, tumor staging relied on the overall tumor size. With the recognition of the lepidic pattern as a noninvasive component in non-mucinous adenocarcinoma, the eighth edition shifted staging to focus exclusively on the invasive portion of the tumor.⁶ This paradigm change has been widely implemented in clinical practice, with its utility and prognostic value thoroughly validated.^14,15

The IASLC grading system defines high-grade patterns as solid, micropapillary, and complex glandular (cribriform and fused glands) patterns.⁸ The complex glandular pattern includes cribriform structures and fused glands. These patterns are associated with high mitotic rates, tumor necrosis, and lymphovascular invasion, underscoring their aggressive nature.^16,17 This grading system was proposed based on a multi-institutional retrospective analysis involving independent training and validation cohorts, which established a 20% cutoff for high-grade patterns, demonstrating the best sensitivity and specificity for prognostic predictions.⁸ This system refines prognostic stratification, offering a reproducible and practical tool for clinical pathology.⁸ As a result, the fifth WHO Classification of Thoracic Tumours has recognized this grading system in the section on invasive non-mucinous adenocarcinoma,⁷ and it has also been incorporated into the CAP cancer protocols for lung tumors,¹⁸ further supporting its clinical applicability.

Furthermore, when cases in our study were reclassified according to the proposed grading system, grade 1 tumors, characterized by bland pneumocytic cells growing along the surface of alveolar walls (lepidic predominant tumors), exhibited a mean lepidic component of 76.0 ± 17.3%. Grade 2 tumors, composed of acinar-predominant tumors forming glandular structures and papillary-predominant tumors with glandular tumor cells lining fibrovascular cores, had a mean acinar/papillary component of 71.3 ± 14.7%. In grade 3 tumors, which contained at least 20% of high-grade patterns (solid, micropapillary, and complex glandular structures), the mean high-grade pattern composition was 47.8 ± 26.3%. These findings further validate the rationale behind the 20% cutoff for high-grade patterns, demonstrating its ability to stratify tumors based on their histologic aggressiveness.

Stage IA grade 3 tumors exhibited survival outcomes comparable to stage IB, underscoring their high-risk nature. Given that current guidelines differentiate treatment strategies between stage IA and IB,¹⁹ these findings suggest that histologic grade 3 may have clinical implications in refining risk stratification for early stage lung adenocarcinoma. When compared with pleural invasion, a well-established risk factor incorporated into staging systems, histologic grade 3 demonstrated a potentially greater prognostic impact in certain aspects. Considering the comparable survival outcomes between stage IA grade 3 and stage IB tumors, it may be beneficial to evaluate the role of adjuvant therapy in these patients, aligning treatment strategies more closely with those for stage IB disease.

To account for the potential interobserver variability associated with the histologic grading system, particularly concerning the 20% cutoff for high-grade patterns, survival analyses were also performed using an alternative grouping based on the presence of any high-grade patterns (>0%). Tumors with any high-grade patterns demonstrated significantly worse outcomes compared with those without, consistent with the prognostic implications of high-grade histology. However, DFS analysis revealed that stage IA tumors with any high-grade patterns maintained a significant survival difference from stage IB tumors (Supplementary Fig. 2).

These findings suggest that the 20% cutoff for grade 3 provides stronger risk stratification power, as it identifies tumors with prognostic risks comparable to stage IB. In contrast, grouping based on any high-grade patterns (>0%) seems to have less discriminative power for risk stratification, particularly when considering DFS outcomes. This supports the robustness of the current 20% cutoff in identifying high-risk groups while emphasizing the need for continued standardization to minimize interobserver variability in clinical practice. Notably, recent studies have reported good to excellent interobserver agreement for histologic grading systems, with agreement levels ranging from 0.61 to 0.94 in various studies.^8,20

This study has inherent limitations that warrant consideration. As a single-center, retrospective analysis, it is susceptible to biases inherent in such study designs, including selection bias and potential data imbalances. The absence of a dedicated validation cohort limits the ability to generalize these findings across broader populations. Furthermore, although the follow-up period exceeded five years, the retrospective nature of the study precludes accounting for factors such as cause-specific mortality or recurrence patterns, which could further elucidate survival differences observed among the subgroups. In addition, the study design does not permit the assessment of potential biases such as lead-time bias, which might influence interpretations of risk stratification within early stage disease.

Despite these limitations, this study provides valuable insights into the prognostic significance of histologic grade 3 and its potential role in refining risk stratification in stage I non-mucinous adenocarcinoma. The findings underscore the need for future multicenter, prospective studies to validate and expand upon these results, ensuring broader applicability and further enhancing the understanding of high-grade patterns' clinical implications.

Stage IA tumors represent a heterogeneous group, meeting size criteria without pleural invasion or main bronchus involvement. The newly proposed and recently validated grading system has emerged as the “new normal” in histologic evaluation, offering robust prognostic insights.²¹ Our findings strongly support the integration of histologic grade 3 into the staging system for early stage lung adenocarcinoma. Given its significant prognostic implications, the incorporation of histologic grade into staging criteria could lead to more accurate risk stratification and improved treatment strategies. Although additional studies are needed to establish universal grading standards, our results provide substantial evidence that histologic grading should be considered in future refinements of the TNM classification.

CRediT Authorship Contribution Statement

Gyerim Park: Writing – original draft, Formal analysis.

Youngkyu Moon: Methodology, Supervision, Data curation, Writing – review & editing.

Meejeong Kim: Investigation, Formal analysis, Visualization, Writing – review & editing.

Kyo Young Lee: Methodology, Supervision, Investigation, Writing – review & editing.

Yeoun Eun Sung: Conceptualization, Methodology, Investigation, Visualization, Writing – review & editing.

Disclosure

The authors declare no conflict of interest.