Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jul 1.
Published in final edited form as: Mod Pathol. 2024 May 21;37(7):100520. doi: 10.1016/j.modpat.2024.100520

Novel insights into the International Association for the Study of Lung Cancer grading system for lung adenocarcinoma

Kay See Tan a, Allison Reiner a, Katsura Emoto b, Takashi Eguchi c, Yusuke Takahashi d, Rania G Aly e, Natasha Rekhtman e, Prasad S Adusumilli f,g, William D Travis e
PMCID: PMC11260232  NIHMSID: NIHMS2003976  PMID: 38777035

Abstract

The new grading system for lung adenocarcinoma (ADC) proposed by the International Association for the Study of Lung Cancer (IASLC) defines prognostic subgroups based on histologic patterns observed on surgical specimens. This study seeks to provide novel insights into the IASLC grading system, with a particular focus on recurrence–specific survival (RSS) and lung cancer–specific survival (LCSS) among patients with stage I ADC. Under the IASLC grading system, tumors were classified as grade 1 (lepidic predominant with <20% high-grade patterns [micropapillary, solid, complex glandular]), grade 2 (acinar or papillary predominant with <20% high-grade patterns), or grade 3 (≥20% high-grade patterns). Kaplan-Meier survival estimates, pathologic features, and genomic profiles were investigated for patients whose disease was reclassified to a higher grade under the IASLC grading system on the basis of the hypothesis that they would strongly resemble patients with predominant high-grade tumors. Overall, 423 of 1443 patients (29%) with grade 1 or 2 tumors by the predominant pattern–based grading system had their tumors upgraded to grade 3 by the IASLC grading system. The RSS curves for patients with upgraded tumors were significantly different from those for patients with grade 1 or 2 tumors (log-rank p<0.001) but not from those for patients with predominant high-grade patterns (p=0.3). Patients with upgraded tumors had a similar incidence of visceral pleural invasion and spread of tumor through air spaces as patients with predominant high-grade patterns. In multivariable models, the IASLC grading system remained significantly associated with RSS and LCSS after adjustment for aggressive pathologic features such as visceral pleural invasion and spread of tumor through air spaces. The IASLC grading system outperforms the predominant pattern–based grading system and appropriately reclassifies tumors into higher grades with worse prognosis, even after other pathologic features of aggressiveness are considered.

Keywords: Nonmucinous lung adenocarcinoma, pathological tumor grading, cribriform, spread of tumor through air spaces, prediction model

Introduction

According to the 2015 and 2021 World Health Organization lung tumor classifications,1, 2 invasive nonmucinous adenocarcinoma (ADC) can be categorized as lepidic, papillary, acinar, micropapillary, or solid predominant. A variety of adenocarcinoma grading systems have been proposed including 1) predominant pattern1, 3–6 2) predominant and secondary pattern,7 3) high grade patterns,8, 9 or 4) nuclear grade10 and 5) mitotic grade.11–13 The often-used predominant grading system mentioned in the 2015 WHO Classification stratifies tumors into three prognostic groups on the basis of the predominant pattern.1, 3–6, 14 A drawback of the predominant pattern–based grading system is that it does not account for the potential prognostic impact of nonpredominant high-grade patterns. Studies have demonstrated that even a minor percentage of micropapillary or solid pattern is associated with poor prognosis, particularly in stage I lung ADC.14–17 In addition to the five major histologic patterns identified in the 2015 and 2021 WHO classifications,1, 2 cribriform and fused glands have emerged as high-risk histologic patterns associated with poor survival.18–22

On the basis of survival data from a multi-institutional study,23 the International Association for the Study of Lung Cancer (IASLC) Pathology Committee proposed a novel grading system for invasive ADC that incorporates both predominant and nonpredominant high-grade patterns. The IASLC grading system has since been validated for overall survival and progression-free survival in multiple studies.24–28 However, the validity of this grading system for predicting recurrence-specific survival (RSS) and lung cancer–specific survival (LCSS) is not known. These cancer-specific outcomes, in addition to overall survival, are pertinent for patients with stage I lung ADC. The present study, which focuses on patients with stage I invasive nonmucinous lung ADC, seeks to provide novel insights into the IASLC grading system, with a particular focus on RSS and LCSS. In addition, we aim to elucidate pathologically aggressive characteristics of tumors that were reclassified as high grade under the IASLC grading system and demonstrate the prognostic value of the IASLC grading system in the presence of aggressive pathologic features such as visceral pleural invasion (VPI) and spread of tumor through air spaces (STAS), both of which have yet to be investigated in this context. Furthermore, we critically assess the value of two recently proposed modifications to the IASLC grading system.

Materials and Methods

Study Cohort

This retrospective study includes patients with pathologically confirmed stage IA and IB lung ADC who underwent surgical resection at Memorial Sloan Kettering Cancer Center from 1995 to 2014. Patients with ADC in situ, minimally invasive ADC, and special variants, including invasive mucinous ADC and colloid ADC, were excluded from the study cohort. Clinical and survival data were collected from the prospectively maintained Thoracic Surgery Service lung ADC database and from review of patient medical records. Pathologic stage was assigned on the basis of the eighth edition of the American Joint Committee on Cancer TNM Staging Manual.29 This study was approved by the institutional review board at Memorial Sloan Kettering Cancer Center, which included a waiver of informed consent.

Histologic Evaluation and Grading Criteria

Details regarding the histologic evaluation of this study cohort have been described previously.30 All hematoxylin and eosin–stained slides from resected tumors were reviewed by two pathologists (K.E. and W.D.T.) using an Olympus BX51 microscope (Olympus, Tokyo, Japan) with a standard 22-mm-diameter eyepiece. Both pathologists were blinded to the clinical data of the patients, and any discrepancies in histologic evaluation between the pathologists were resolved by consensus using a multihead microscope. Quantification of each histologic pattern (lepidic, acinar, papillary, micropapillary, and solid) was recorded in increments of 5%. After the publication of the IASLC study,18 cribriform was also quantified as a high-grade histologic pattern separate from acinar subtype. For each tumor, the predominant pattern was defined as the pattern with the largest percentage, even if the pattern was <50% of the tumor. VPI, lymphovascular invasion (LVI), STAS, and necrosis were also recorded as absent or present.

Tumors from each patient were classified into grades using two grading systems. First, using the predominant pattern–based grading system, tumors were classified according to previously established conventions: grade 1 (lepidic predominant), grade 2 (acinar or papillary predominant), and grade 3 (micropapillary or solid predominant). Next, on the basis of the proposed IASLC grading system, tumors were classified as 1 of 3 grades: grade 1 (lepidic predominant with no or <20% high-grade patterns), grade 2 (acinar or papillary predominant with no or <20% high-grade patterns), or grade 3 (any tumor with ≥20% high-grade patterns). In the IASLC classification, high-grade patterns include micropapillary, solid, and complex glandular patterns (cribriform and fused glands). Grade 1 or 2 tumors by the predominant pattern–based grading system that were reclassified to grade 3 by the IASLC grading system were considered to be upgraded.

The study also considered two alternative approaches that have been introduced since the adoption of the IASLC grading system. In the first, Park et al.31 proposed to subdivide IASLC grade 2 tumors (acinar or papillary predominant with <20% high-grade patterns) into two groups on the basis of lepidic component: ≥10% (grade 2A) and <10% (grade 2B). In the second proposal, Bosse et al.28 advocated for a “simplified grading system” that removes complex glandular patterns from the high-grade pattern definition of the IASLC grading system. The authors asserted that, since the simplified version performed as well as the IASLC grading system, the preference should be for the simpler grading system, as it requires fewer resources, less expertise, and a smaller burden on pathologists. Thus, in their system, grade 3 will include only tumors with predominant or ≥20% of micropapillary and solid patterns, without accounting for cribriform or fused glands.

Recurrence and Follow-up

The recurrence definitions have been previously described.15 All recurrences were confirmed by cytologic or histologic evaluation after clinical and/or radiologic suspicion. Recurrences were categorized in accordance with the Society of Thoracic Surgeons Workforce recommendations32: (i) local recurrence was defined by evidence of a tumor in the same lobe or at the surgical margin of the original tumor; (ii) regional recurrence was defined by evidence of a tumor in a second ipsilateral lobe, in the ipsilateral hilar lymph nodes (N1), or in the ipsilateral mediastinal lymph nodes (N2); and (iii) distant recurrence was defined by evidence of a tumor in the contralateral lung, in the contralateral mediastinal or ipsilateral supraclavicular lymph nodes (N3), or elsewhere outside the hemithorax. Recurrence in this study refers to the combined local, regional, and distant recurrence events as defined. All patients were evaluated postoperatively with chest x-ray, chest computed tomography scan, and positron emission tomography scan, when clinically indicated, in addition to periodic clinical follow-up, in accordance with National Comprehensive Cancer Network guidelines.33

Statistical Analysis

Patient characteristics were summarized as frequency and percentages for categorical variables and as median and interquartile ranges for continuous variables. The distributions of grades between the two grading systems were compared using the Bhapkar test34 to assess whether applying the IASLC grading system resulted in a significant shift in the distribution of grades. Clinical, pathologic, and genomic features were summarized for patients with upgraded tumors compared against those with predominant high-grade patterns using the Wilcoxon rank-sum test for continuous factors and the chi-squared test for categorical factors. Comparisons between the three IASLC grades were conducted using the Kruskal-Wallis test and chi-squared test.

RSS was calculated using the Kaplan-Meier approach from the time of surgery to the time of recurrence and was otherwise censored at the time of last follow-up or death. LCSS was calculated from the time of surgery to the time of death from lung cancer or was otherwise censored at the time of last follow-up or death. Survival was compared between grades using log-rank tests. As a sensitivity analysis, the cumulative incidence of recurrence was summarized, with death without recurrence treated as a competing event, as was the cumulative incidence of lung cancer death, with death from other causes treated as a competing event. Cumulative incidence curves were compared between grades using Gray’s test. Other survival endpoints were calculated from the time of surgery to the time of recurrence or death (for recurrence-free survival) or the time of death (for overall survival). The median duration of follow-up was derived using the reverse Kaplan-Meier approach.

The performance of each grading system was quantified by C-index (for discrimination performance), calibration curves (for calibration performance), and the Akaike information criterion (AIC), on the basis of multivariable Cox models with adjustment for the same set of factors in the original grading system study (age, sex, surgery type, and pathologic stage IA or IB).18 Additional clinical factors were considered in the multivariable models: pleural invasion, LVI, STAS, and necrosis. Statistical analyses were conducted using R 3.5.1 (R Development Core, Vienna, Austria). All significance tests were two-sided, and p<0.05 was set as the level of statistical significance.

Results

Summary of Patient Characteristics

Overall, 1443 patients were included: 77% had stage IA disease, 43% had LVI, and 45% had STAS (Table 1). The median duration of follow-up was 5.6 years (interquartile range, 3.3-8.1 years). In total, 229 patients had recurrence, and 391 died (134 of lung cancer). The most prevalent pattern present at ≥5% was acinar (92%), followed by papillary (73%) and lepidic (60%). On the basis of the 2015 and 2021 WHO definitions, the most prevalent predominant subtype was acinar ADC (46%), followed by papillary (17%), solid (16%), lepidic (11%), and micropapillary (6%).

Table 1.

Patient characteristics (N=1443)

 Characteristic Total
Age 70 (62-76)
Sex
  Female 882 (61)
  Male 561 (39)
Smoking status
  Never 251 (17)
  Former or current 1192 (83)
Surgery type
  Pneumonectomy, bilobectomy, or lobectomy 1027 (71)
  Segmentectomy or wedge resection 416 (29)
Mutation status
  Wild-type 618 (51)
  EGFR 248 (20)
  KRAS 349 (29)
  Unknown 228
Pathologic stage
  IA1 321 (22)
  IA2 575 (40)
  IA3 211 (15)
  IB 336 (23)
Gross tumor size, cm 1.80 (1.40-2.50)
Invasive tumor size, cm 1.50 (1.04-2.19)
Visceral pleural invasion 250 (17)
Lymphovascular invasion 623 (43)
Necrosis 213 (15)
Spread through air spaces 638 (45)
  Unknown 13
Predominant subtype
  Lepidic 152 (11)
  Acinar 666 (46)
  Papillary 251 (17)
  Micropapillary 144 (10.0)
  Solid 230 (16)
Lepidic present 866 (60)
Acinar present 1327 (92)
Papillary present 1060 (73)
Solid present 653 (45)
Micropapillary present 800 (55)
Adjuvant chemotherapy
  No 1288 (89)
  Yes 32 (2.2)
  Unknown 123 (8.5)
IASLC grade
  1 139 (9.6)
  2 507 (35)
  3 797 (55)
Predominant pattern–based grade
  1 152 (11)
  2 917 (64)
  3 374 (26)
Open in a new tab

Data are median (interquartile range) or no. (%). IASLC, International Association for the Study of Lung Cancer.

Survival Outcomes Based on the Grading Systems

According to the predominant pattern–based grading system, 11% of patients (152/1443) had grade 1 tumors, 64% (917/1443) had grade 2 tumors, and 26% (374/1443) had grade 3 tumors. The corresponding proportions using the IASLC grading system were 9.6% grade 1, 35% grade 2, and 55% grade 3 (Table 1).

RSS was significantly different between the three grades when the predominant pattern–based grading system was used (log-rank p<0.001; Figure 1) and when the IASLC grading system was used (p<0.001; Figure 1). Similarly, LCSS was significantly different between the three grades when the predominant pattern–based grading system was used (p<0.001; Figure 1) and when the IASLC grading system was used (p<0.001; Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Recurrence-specific survival and lung cancer–specific survival using the predominant pattern–based grading system (A-B) and the International Association for the Study of Lung Cancer (IASLC) grading system (C-D). Survival patterns of patients whose tumors were upgraded to grade 3 by the IASLC grading system (E-F). Patients with upgraded tumors (black dashed lines) are those with predominant lepidic, acinar, or papillary patterns but that include ≥20% high-grade patterns.

Reclassification of Tumors Using the IASLC Grading System

Overall, 423 patients (29%) were considered to have had their tumors upgraded: 13 grade 1 tumors by the predominant pattern–based grading system were upgraded to grade 3 by the IASLC grading system, and 410 grade 2 tumors by the predominant pattern–based grading system were upgraded to grade 3 by the IASLC grading system (Figure 2). The reclassification of tumors resulted in a significant shift in grade distribution between the two grading systems (Bhapkar test p<0.001). These patterns in reclassification were consistent with those in recent papers that reported summary data using the IASLC grading system (Supplementary Table 1).

Figure 2.

Figure 2.

Open in a new tab

Reclassification from predominant pattern-based grading system to IASLC grading system. Red box: 423 of 1443 patients (29%) with grade 1 or grade 2 tumors by the predominant pattern-based grading system had their tumors upgraded to grade 3 by the IASLC grading system.

The RSS curve for patients with upgraded tumors closely tracked the curve for patients with predominant high-grade patterns (Figure 1E). The RSS curves for patients with upgraded tumors were significantly different from those for patients with grade 1 (p<0.001) or 2 (p<0.001) tumors but were not significantly different from those for patients with predominant high-grade patterns (p=0.3). The LCSS curve for patients with upgraded tumors did not track as closely to the curve for patients with predominant high-grade patterns as was observed for RSS (Figure 1F), but it did remain disparate from the curve for patients with grade 1 or 2 tumors. The survival curves for patients with upgraded tumors were significantly different from those for patients with grade 1 (p=0.003), grade 2 (p<0.001), or predominant high-grade pattern (p=0.03 1) tumors.

Similar observations were noted for cumulative incidence of recurrence, cumulative incidence of lung cancer deaths, recurrence-free survival, and overall survival (Figure 3, Supplementary Figure 1).

Figure 3.

Figure 3.

Open in a new tab

Cumulative incidence of recurrence and lung cancer deaths using the predominant pattern–based grading system (A-B) and the International Association for the Study of Lung Cancer (IASLC) grading system (C-D). Survival patterns of patients whose tumors were upgraded to grade 3 by the IASLC grading system (E-F). Patients with upgraded tumors (black dashed lines) are those with predominant lepidic, acinar, or papillary patterns but that include ≥20% high-grade patterns.

Clinical Features of Patients with Upgraded Tumors

Compared with patients with predominant high-grade patterns, patients with upgraded tumors were more likely to be never-smokers (17% vs 9.1%; p=0.002), have an EGFR mutation (17% vs 9.3%; p=0.018), and have stage IA disease (73% vs 65%; p=0.017) (Supplementary Table 2). Rates of VPI and STAS were not significantly different between the two groups. In contrast, patients with upgraded tumors were less likely to have LVI (56% vs 68%; p<0.001) or necrosis (17% vs 31%; p<0.001). Overall, patients with upgraded tumors were more likely to have aggressive pathologic features, compared with patients with grade 1 or 2 tumors, and rates of aggressive pathologic features among patients with upgraded tumors were similar to rates among patients with predominant high-grade patterns (Table 2). Last, the percentage of acinar component at the tumor-level was higher among patients with upgraded tumors (median, 35% vs 10%; p<0.001), whereas the percentage of solid component was lower among patients with upgraded tumors (median, 10% vs 50%; p<0.001) (Table 2 and Supplementary Table 2).

Table 2.

Patient characteristics by the IASLC grading system

Characteristic IASLC Grades 1-3
 IASLC Grade 3 (Subset by Predominant High-Grade Pattern or Not)
Grade 1 (n=139) Grade 2 (n=507) Grade 3 (n=797) p Valuea Predominant High-Grade Pattern (n=374) Upgraded (n=423) p Valueb
Age 71 (66-76) 70 (62-76) 69 (62-75) 0.13 69 (62-75) 70 (62-76) 0.4
Sex 0.4 0.6
  Female 81 (58) 322 (64) 479 (60) 221 (59) 258 (61)
  Male 58 (42) 185 (36) 318 (40) 153 (41) 165 (39)
Smoking status <0.001 0.002
  Never 31 (22) 116 (23) 104 (13) 34 (9.1) 70 (17)
  Ever 108 (78) 391 (77) 693 (87) 340 (91) 353 (83)
Mutation status <0.001 0.018
  Wild-type 54 (47) 195 (45) 369 (55) 170 (57) 199 (54)
  EGFR 36 (31) 123 (28) 89 (13) 28 (9.3) 61 (17)
  KRAS 26 (22) 114 (26) 209 (31) 102 (34) 107 (29)
  Unknown 23 75 130 74 56
Gross tumor size, cm 1.80 (1.50-2.50) 1.80 (1.20-2.50) 1.80 (1.40-2.50) 0.14 1.80 (1.40-2.50) 1.80 (1.40-2.50) >0.9
Pathologic stage <0.001 0.017
  IA 133 (96) 425 (84) 549 (69) 242 (65) 307 (73)
  IB 6 (4.3) 82 (16) 248 (31) 132 (35) 116 (27)
Visceral pleural invasion 6 (4.3) 53 (10) 191 (24) <0.001 98 (26) 93 (22) 0.2
Lymphovascular invasion 9 (6.5) 126 (25) 488 (61) <0.001 253 (68) 235 (56) <0.001
Necrosis 4 (2.9) 22 (4.3) 187 (23) <0.001 115 (31) 72 (17) <0.001
Spread through air spaces 17 (12) 136 (27) 485 (61) <0.001 237 (64) 248 (59) 0.2
  Unknown 1 6 6 3 3
Pattern, %
  Lepidic 60 (50-63) 20 (10-30) 0 (0-10) <0.001 0 (0-5) 5 (0-10) <0.001
  Acinar 30 (20-35) 50 (30-60) 20 (10-40) <0.001 10 (5-20) 35 (20-50) <0.001
  Papillary 10 (0-20) 25 (10-45) 10 (0-20) <0.001 5 (0-10) 20 (5-30) <0.001
  Micropapillary 0 (0-5) 0 (0-5) 10 (0-25) <0.001 10 (0-45) 10 (0-20) 0.003
  Solid 0 (0-0) 0 (0-0) 20 (0-50) <0.001 50 (10-80) 10 (0-20) <0.001
  Complex glandular patterns 0 (0-0) 0 (0-0) 0 (0-10) <0.001 0 (0-5) 0 (0-20) <0.001
  Classical micropapillary 0 (0-0) 0 (0-5) 5 (0-10) <0.001 5 (0-15) 5 (0-10) 0.050
  Filigree micropapillary 0 (0-0) 0 (0-0) 0 (0-10) <0.001 5 (0-20) 0 (0-10) <0.001
Open in a new tab

Data are median (interquartile range) or no. (%). IASLC, International Association for the Study of Lung Cancer.

a

Kruskal-Wallis rank-sum test; Pearson’s chi-squared test.

b

Wilcoxon rank-sum test; Pearson’s chi-squared test.

Performance of the Grading Systems

After adjustment for baseline covariates used in the initial proposal,23 both the predominant pattern–based system and the IASLC grading system were associated with RSS (p<0.001; Supplementary Table 3) and LCSS (p<0.001; Supplementary Table 4). On the basis of these multivariable models, discriminative performance was similar between the predominant pattern–based grading system (C-index RSS, 0.70; C-index LCSS, 0.74) and the IASLC grading system (C-index RSS, 0.72; C-index LCSS, 0.74). Calibration curves indicate good calibration performance across all outcomes for both grading systems (Supplementary Figure 2).

Relationship Between the IASLC Grading System and Pathologic and Genomic Features

In the presence of additional pathologic features, the IASLC grading system remained associated with RSS (Supplementary Table 5). STAS, LVI, and necrosis were independently associated with RSS in the multivariable models. Of note, VPI was not significantly associated with the survival outcomes in the presence of LVI. When summarized by the component of high-grade patterns, the proportion of patients with STAS increased as the component of high-grade patterns increased, until 20% presence of high-grade patterns, at which point the proportion of patients with STAS began to plateau (Supplementary Figure 3A). Among patients with genomic data (n=1215), as the component of high-grade patterns increased, the proportion of patients with EGFR mutation decreased, whereas the proportion of patients with KRAS mutation remained stable (Supplementary Figure 3B).

Validation of Two Proposed Modified Grading Systems

Last, the study data were applied to two recently proposed modifications to the IASLC grading system.28, 31 When patients with grade 2 tumors were subdivided into two groups on the basis of lepidic component, RSS and LCSS were not statistically significantly different between those with ≥10% lepidic component and those with <10% lepidic component (Figure 4A).

Figure 4.

Figure 4.

Open in a new tab

Recurrence-specific survival and lung cancer–specific survival for the proposed modifications to the International Association for the Study of Lung Cancer grading system. (A) Proposal by Park et al. to subdivide grade 2 into tumors with ≥10% versus <10% lepidic pattern. (B) Proposal by Bosse et al. to exclude complex glandular patterns as a high-grade pattern. (C) confirmation of the value of complex glandular patterns as a high-grade pattern. The orange solid lines include patients whose tumors would otherwise have remained grade 2 by the International Association for the Study of Lung Cancer grading system if complex glandular patterns were excluded as a high-grade pattern. Recurrence-specific survival for this group (orange solid line) was significantly different from that for patients with grade 1 or 2 tumors (p<0.001) but was not significantly different from that for patients with grade 3 tumors (p=0.4). Lung cancer–specific survival for this group was significantly different from that for patients with grade 1 (p=0.005) or 2 (p=0.012) tumors but was not significantly different from that for patients with grade 3 tumors (p=0.4). ACI, acinar; CGP, complex glandular patterns; LPD, lepidic; MIP, micropapillary; PAP, papillary; SOL, solid.

When the simplified IASLC grading system was applied, both RSS and LCSS were significantly different between the three grades (Figure 4B). The discriminative performance of the simplified grading system (AUC=0.716) was similar to that of the IASLC grading system (AUC=0.718). Additionally, the AIC of the simplified grading system was higher than that of the IASLC grading system (AIC=3080 vs 3071), indicating that the IASLC grading system was superior in terms of model fit. In addition to C-index and AIC, the value of complex glandular patterns as a high-grade pattern can be evaluated by focusing on the subset of patients with tumor micropapillary and solid components below the 20% threshold but classified as IASLC grade 3 because of the inclusion of complex glandular patterns. RSS and LCSS for this subset of patients were significantly different from those for patients with grade 1 or 2 tumors (RSS: subset vs grade 1, log-rank p<0.001; subset vs grade 2, p=0.001) but were not significantly different from those from patients with grade 3 tumors (RSS: subset vs grade 3, p=0.4) (Figure 4C).

Discussion

In a large cohort of patients with stage I lung ADC, this study is the first, to our knowledge, to (1) demonstrate the improvement of the IASLC grading system over the predominant pattern–based grading system by focusing on RFS and LCSS, which are pertinent to stage I lung ADC; (2) establish the association of known aggressive pathologic characteristics in patients with upgraded tumors; and (3) perform an analysis of outcomes after the inclusion of aggressive pathologic features, such as VPI and STAS. In addition, we have highlighted that the proportions of patients with grade 3 tumors vary across cohorts from different institutions and countries (Supplementary Table 1), providing ground for future research.

The IASLC grading system was mentioned by the 2021 World Health Organization Classification of Thoracic Tumors35 and is the first officially recognized grading system for invasive nonmucinous lung ADC. However, this proposal was based on a single study from the IASLC Pathology Committee,23 and further validation was needed. The present study is the first study to demonstrate the improvement of the IASLC grading system over the predominant pattern–based grading system by focusing on the clinical features and cancer-specific prognoses of patients with upgraded tumors. Tumors that were reclassified to grade 3 were distinctive from grade 1 and 2 tumors and were similar to tumors with predominant high-grade patterns in terms of poor prognostic features. Most importantly, the survival patterns of patients with upgraded tumors closely tracked those of patients with predominant high-grade patterns in terms of recurrence, lung cancer–specific deaths, and overall survival. By focusing on the survival patterns of these patients with upgraded tumors, we link the impact of the proposed changes in the IASLC grading system to actual patient outcomes. Whereas most studies thus far have used the C-index to quantify the performance of the IASLC grading system, there is no direct link between C-index (or any other discrimination measure) and clinical outcomes.36–38 Hence, two models can provide similar C-indices despite assigning two different risk classifications for the same patient. Instead, the current approach to isolate and study the features of these patients with upgraded tumors revealed observations that would not have been evident using traditional performance metrics, such as C-index.

This study also investigated two recent studies that proposed modifications to the IASLC grading system. Results using our data did not validate either proposal.28, 31 By use of the first proposal, our data show that recurrence and lung cancer–specific deaths are not statistically significantly different between patients with grade 2 tumors with ≥10% lepidic component and <10% lepidic component.31 Therefore, the question remains whether the percentage of lepidic component should be incorporated into the grading system for grade 2 tumors. It is important to note that this proposed incorporation of lepidic component was first studied in an Asian cohort, which was motivated by the well-documented differences in patient populations between Asian and Western cohorts. For example, even after reclassification, the proportion of patients with IASLC grade 2 tumors was higher among Asian than non-Asian cohorts (Supplementary Table 1). With the smaller proportion of patients with grade 2 tumors in our cohort, subdividing this group may not retain adequate statistical power to detect significant differences.

In the second proposal, Bosse et al.28 advocate for a simplified version of the IASLC grading system that considers only micropapillary and solid components as high-grade patterns. However, our data did not validate their findings28: although the discrimination performance was similar between the two grading systems, our data demonstrate that the IASLC grading system was superior to the simplified version in terms of model fit. Furthermore, the contribution of complex glandular patterns as a high-grade feature was confirmed in our data. Specifically, patients with tumors that would have remained as grade 1 or 2 by not including cribriform or fused glands as high-grade patterns had survival outcomes that were similar to those for patients with predominant high-grade patterns and were significantly different from those for patients with grade 1 or 2 tumors. Hence, the evidence demonstrates that it is still preferable to include cribriform and fused glands as high-grade patterns. Additionally, whereas Bosse et al.28 assessed only overall survival by use of the simplified grading system, we evaluated RSS and LCSS. Further studies will be necessary to address the value of complex glandular patterns in the IASLC grading system among different populations from diverse geographical regions.

Observations from the present study should be contextualized within the body of recent literature that has summarized data using the IASLC grading system (Supplementary Table 1). In the current study, the proportion of patients with grade 3 tumors was 55%, whereas this proportion ranged from 9% to 72% in other studies that have reported data on patients with stage I disease. Asian cohorts tended to have lower proportions of grade 3 tumors than non-Asian cohorts. Relatedly, the proportion of patients with grade 1 or 2 tumors by the predominant pattern-based grading system that were upgraded to grade 3 by the IASLC grading system was 29% in the current study, whereas this proportion ranged from 5% to 39% in other studies. Most interestingly, the study from Bosse et al.28 had the largest proportions, across all studies that reported data on patients with stage I disease, of patients with grade 3 tumors (72%) and patients with grade 2 tumors that were reclassified to grade 3 (67%).

The current study differs from the original IASLC grading system analysis23 in several ways. Instead of using the date of diagnosis as the start time for time-to-event endpoints, the present study used date of surgery. Additionally, the present study investigated LCSS and RSS, both of which were not assessed in the original proposal but are highly relevant for early-stage lung cancer. Last, rather than including all stages, the current study focused only on stage I disease; this is a more homogeneous group of patients for whom pathologic grading may play a substantial role in adjuvant treatment. As a result, observations in this study may not be generalizable to patients with higher-stage disease.

Conclusion

The IASLC grading system outperforms the predominant pattern–based grading system and correctly reclassifies patients with poorer prognosis into the high-grade category. With the confirmation of the prognostic value of this system, future studies and clinical trials should incorporate the pathologic characteristics proposed in the IASLC grading system.

Supplementary Material

1

Acknowledgments

This research was funded, in part, by the National Institutes of Health/National Cancer Institute Cancer Center Support Grant P30 CA008748.

Funding and Role of the Funding Source:

P.S.A.’s laboratory work is supported by grants from the National Institutes of Health (grant numbers P30 CA008748, R01 CA236615-01, R01 CA235667, and U01 CA214195); the U.S. Department of Defense (grant numbers CA170630, CA180889, and CA200437); the Baker Street Foundation; the Batishwa Fellowship; the Comedy versus Cancer Foundation, the Cycle for Survival Fund, the DallePezze Foundation; the Derfner Foundation; the Esophageal Cancer Education Fund; the Geoffrey Beene Foundation; the Memorial Sloan Kettering Technology Development Fund; the Miner Fund for Mesothelioma Research; the Mr. William H. Goodwin and Alice Goodwin, the Commonwealth Foundation for Cancer Research; and the Experimental Therapeutics Center of Memorial Sloan Kettering Cancer Center. P.S.A.’s laboratory received research support from ATARA Biotherapeutics and Novocure Inc. The research support sources did not have any role in study design, collection, analysis, and interpretation of data, writing of the article, or the decision to submit the article for publication.

Abbreviations:

ACI

acinar

ADC

adenocarcinoma

AIC

Akaike information criterion

CGP

complex glandular patterns

IASLC

International Association for the Study of Lung Cancer

LCSS

lung cancer–specific survival

LPD

lepidic

LVI

lymphovascular invasion

MIP

micropapillary

PAP

papillary

RSS

recurrence-specific survival

SOL

solid

STAS

spread of tumor through air spaces

VPI

visceral pleural invasion

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Ethics Approval / Consent to Participate: This study was approved by the institutional review board at Memorial Sloan Kettering Cancer Center, which included a waiver of informed consent.

Conflicts of Interest: All authors have no conflicts of interest to disclose.

Data availability Statement:

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

References

  • 1.Travis WD, Brambilla E, Nicholson AG, et al. The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. Journal of thoracic oncology. 2015;10(9):1243–1260. [DOI] [PubMed] [Google Scholar]
  • 2.World Health Organization. Classification of Tumours. Thoracic Tumours (M). IARC Press: Lyon, France; 2021. [Google Scholar]
  • 3.Boland JM, Wampfler JA, Yang P, Yi ES. Growth pattern-based grading of pulmonary adenocarcinoma-Analysis of 534 cases with comparison between observers and survival analysis. Lung Cancer. 2017;109:14–20. 10.1016/j.lungcan.2017.04.013. [DOI] [PubMed] [Google Scholar]
  • 4.Travis WD, Brambilla E, Geisinger KR. Histological grading in lung cancer: one system for all or separate systems for each histological type? Eur Respir J. 2016;47(3):720–723. 10.1183/13993003.00035-2016. [DOI] [PubMed] [Google Scholar]
  • 5.Zhao ZR, Xi SY, Li W, et al. Prognostic impact of pattern-based grading system by the new IASLC/ATS/ERS classification in Asian patients with stage I lung adenocarcinoma. Lung Cancer. 2015;90(3):604–609. 10.1016/j.lungcan.2015.10.026. [DOI] [PubMed] [Google Scholar]
  • 6.Zombori T, Furak J, Nyari T, Cserni G, Tiszlavicz L. Evaluation of grading systems in stage I lung adenocarcinomas: a retrospective cohort study. J Clin Pathol. 2018;71(2):135–140. 10.1136/jclinpath-2016-204302. [DOI] [PubMed] [Google Scholar]
  • 7.Sica G, Yoshizawa A, Sima CS, et al. A grading system of lung adenocarcinomas based on histologic pattern is predictive of disease recurrence in stage I tumors. Am J Surg Pathol. 2010;34(8):1155–1162. 10.1097/PAS.0b013e3181e4ee32. [DOI] [PubMed] [Google Scholar]
  • 8.Bertoglio P, Aprile V, Ventura L, et al. Impact of High-Grade Patterns in Early-Stage Lung Adenocarcinoma: A Multicentric Analysis. Lung. 2022;200(5):649–660. 10.1007/s00408-022-00561-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Jeon HW, Kim YD, Sim SB, Moon MH. Comparison of clinical results between high grade patterns in stage I lung adenocarcinoma. Thorac Cancer. 2022;13(17):2473–2479. 10.1111/1759-7714.14578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Nakazato Y, Minami Y, Kobayashi H, et al. Nuclear grading of primary pulmonary adenocarcinomas: correlation between nuclear size and prognosis. Cancer. 2010;116(8):2011–2019. 10.1002/cncr.24948. [DOI] [PubMed] [Google Scholar]
  • 11.Duhig EE, Dettrick A, Godbolt DB, et al. Mitosis trumps T stage and proposed international association for the study of lung cancer/american thoracic society/european respiratory society classification for prognostic value in resected stage 1 lung adenocarcinoma. J Thorac Oncol. 2015;10(4):673–681. 10.1097/JTO.0000000000000446. [DOI] [PubMed] [Google Scholar]
  • 12.Kadota K, Suzuki K, Kachala SS, et al. A grading system combining architectural features and mitotic count predicts recurrence in stage I lung adenocarcinoma. Mod Pathol. 2012;25(8):1117–1127. 10.1038/modpathol.2012.58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.von der Thusen JH, Tham YS, Pattenden H, et al. Prognostic significance of predominant histologic pattern and nuclear grade in resected adenocarcinoma of the lung: potential parameters for a grading system. J Thorac Oncol. 2013;8(1):37–44. 10.1097/JTO.0b013e318276274e. [DOI] [PubMed] [Google Scholar]
  • 14.Yoshizawa A, Motoi N, Riely GJ, et al. Impact of proposed IASLC/ATS/ERS classification of lung adenocarcinoma: prognostic subgroups and implications for further revision of staging based on analysis of 514 stage I cases. Mod Pathol. 2011;24(5):653–664. 10.1038/modpathol.2010.232. [DOI] [PubMed] [Google Scholar]
  • 15.Nitadori J, Bograd AJ, Kadota K, et al. Impact of micropapillary histologic subtype in selecting limited resection vs lobectomy for lung adenocarcinoma of 2cm or smaller. J Natl Cancer Inst. 2013;105(16):1212–1220. 10.1093/jnci/djt166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Warth A, Muley T, Meister M, et al. The novel histologic International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society classification system of lung adenocarcinoma is a stage-independent predictor of survival. Journal of clinical oncology. 2012;30(13):1438–1446. [DOI] [PubMed] [Google Scholar]
  • 17.Takahashi Y, Ishii G, Aokage K, Hishida T, Yoshida J, Nagai K. Distinctive histopathological features of lepidic growth predominant node-negative adenocarcinomas 3-5 cm in size. Lung Cancer. 2013;79(2):118–124. 10.1016/jlungcan.2012.10.013. [DOI] [PubMed] [Google Scholar]
  • 18.Moreira AL, Joubert P, Downey RJ, Rekhtman N. Cribriform and fused glands are patterns of high-grade pulmonary adenocarcinoma. Hum Pathol. 2014;45(2):213–220. 10.1016/j.humpath.2013.10.011. [DOI] [PubMed] [Google Scholar]
  • 19.Kadota K, Kushida Y, Kagawa S, et al. Cribriform subtype is an independent predictor of recurrence and survival after adjustment for the eighth edition of TNM staging system in patients with resected lung adenocarcinoma. Journal of Thoracic Oncology. 2019;14(2):245–254. [DOI] [PubMed] [Google Scholar]
  • 20.Ding Q, Chen D, Wang X, et al. Characterization of lung adenocarcinoma with a cribriform component reveals its association with spread through air spaces and poor outcomes. Lung Cancer. 2019;134:238–244. [DOI] [PubMed] [Google Scholar]
  • 21.Warth A, Muley T, Kossakowski C, et al. Prognostic impact and clinicopathological correlations of the cribriform pattern in pulmonary adenocarcinoma. J Thorac Oncol. 2015;10(4):638–644. 10.1097/JTO.0000000000000490. [DOI] [PubMed] [Google Scholar]
  • 22.Qu Y, Lin H, Zhang C, Li K, Zhang H. Cribriform pattern in lung invasive adenocarcinoma correlates with poor prognosis in a Chinese cohort. Pathol Res Pract. 2019;215(2):347–353. 10.1016/j.prp.2018.12.014. [DOI] [PubMed] [Google Scholar]
  • 23.Moreira AL, Ocampo PS, Xia Y, et al. A grading system for invasive pulmonary adenocarcinoma: a proposal from the International Association for the Study of Lung Cancer Pathology Committee. Journal of Thoracic Oncology. 2020;15(10):1599–1610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kagimoto A, Tsutani Y, Kambara T, et al. Utility of newly proposed grading system from International Association for the Study of Lung Cancer for invasive lung adenocarcinoma. JTO Clinical and Research Reports. 2021;2(2):100126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Rokutan-Kurata M, Yoshizawa A, Ueno K, et al. Validation study of the International Association for the Study of Lung Cancer histologic grading system of invasive lung adenocarcinoma. Journal of Thoracic Oncology. 2021;16(10):1753–1758. [DOI] [PubMed] [Google Scholar]
  • 26.Deng C, Zheng Q, Zhang Y, et al. Validation of the novel International Association for the Study of Lung Cancer grading system for invasive pulmonary adenocarcinoma and association with common driver mutations. Journal of Thoracic Oncology. 2021;16(10):1684–1693. [DOI] [PubMed] [Google Scholar]
  • 27.Hou L, Wang T, Chen D, et al. Prognostic and predictive value of the newly proposed grading system of invasive pulmonary adenocarcinoma in Chinese patients: a retrospective multicohort study. Modern Pathology. 2022;35(6):749–756. [DOI] [PubMed] [Google Scholar]
  • 28.Bossé Y, Gagné A, Althakfi WA, et al. A Simplified Version of the IASLC Grading System for Invasive Pulmonary Adenocarcinomas With Improved Prognosis Discrimination. The American Journal of Surgical Pathology. 2023;47(6):686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Amin MB, Greene FL, Edge SB, et al. The eighth edition AJCC cancer staging manual: continuing to build a bridge from a population-based to a more “personalized” approach to cancer staging. CA: a cancer journal for clinicians. 2017;67(2):93–99. [DOI] [PubMed] [Google Scholar]
  • 30.Emoto K, Eguchi T, Tan KS, et al. Expansion of the concept of micropapillary adenocarcinoma to include a newly recognized filigree pattern as well as the classical pattern based on 1468 stage I lung adenocarcinomas. Journal of Thoracic Oncology. 2019;14(11):1948–1961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Park BJ, Woo W, Cha YJ, et al. Proposal of a revised International Association for the Study of Lung Cancer grading system in pulmonary non-mucinous adenocarcinoma: The importance of the lepidic proportion. Lung Cancer. 2023;175:1–8. 10.1016/j.lungcan.2022.11.003. [DOI] [PubMed] [Google Scholar]
  • 32.Donington J, Ferguson M, Mazzone P, et al. American College of Chest Physicians and Society of Thoracic Surgeons consensus statement for evaluation and management for high-risk patients with stage I non-small cell lung cancer. Chest. 2012;142(6):1620–1635. 10.1378/chest.12-0790. [DOI] [PubMed] [Google Scholar]
  • 33.Ettinger DS, Akerley W, Bepler G, et al. Non-small cell lung cancer. J Natl Compr Canc Netw. 2010;8(7):740–801. 10.6004/jnccn.2010.0056. [DOI] [PubMed] [Google Scholar]
  • 34.Bhapkar VP. A note on the equivalence of two test criteria for hypotheses in categorical data. Journal of the American Statistical Association. 1966;61(313):228–235. [Google Scholar]
  • 35.Thoracic Tumours: International Agency for Research on Cancer; 2021. [Google Scholar]
  • 36.Wey A, Salkowski N, Kasiske BL, et al. The relationship between the C-statistic and the accuracy of program-specific evaluations. American Journal of Transplantation. 2019;19(2):407–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Austin PC, Reeves MJ. The relationship between the C-statistic of a risk-adjustment model and the accuracy of hospital report cards: a Monte Carlo Study. Medical care. 2013;51(3):275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Cook NR. Use and misuse of the receiver operating characteristic curve in risk prediction. Circulation. 2007;115(7):928–935. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2003976-supplement-1.pdf (1.6MB, pdf)

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

RESOURCES