Abstract
This cohort study examines the association between tumor-infiltrating lymphocyte classification and disease progression among patients with metastatic primary cutaneous melanoma receiving checkpoint inhibitor therapy.
Checkpoint inhibitor therapy (CIT) has revolutionized treatment of advanced melanoma and other cancers.1 However, only about 40% of patients respond to CIT, and immune-related adverse effects are common.1 A better ability to predict response could spare likely nonresponders from adverse effects. Elevated tumor mutational burden is associated with increased neoantigens in melanoma, which are in turn associated with increased CD8+ tumor-infiltrating lymphocytes (TILs).2 Prior studies yielded conflicting results regarding a correlation between TILs in primary melanomas of patients with advanced disease and CIT.3,4 We examined the association of TILs classification in primary cutaneous melanoma with CIT response in patients with metastatic disease.
Methods
In this cohort study, we identified patients with stage IV melanoma treated with CIT between January 2012 and December 2017 at the Hospital of the University of Pennsylvania, Philadelphia. Age at therapy initiation, sex, therapy duration and type, and therapeutic response based on Response Evaluation Criteria in Solid Tumors were recorded and coded as progressive disease (PD) or nonprogressive disease (NPD, defined as complete response or stable disease). We located dermatopathology reports for primary cutaneous melanoma and analyzed reports classifying TILs with standard nomenclature: absent; present, nonbrisk; or present, brisk (eMethods in the Supplement). This study was exempted from approval and informed consent by the University of Pennsylvania Institutional Review Board because data were deidentified. We followed the STROBE reporting guideline.
Data were analyzed from February 2019 to February 2022. Univariable and multivariable logistic regression was performed using Stata, version 15. Two-sided P < .05 was significant. Akaike information criterion (AIC) model selection and the likelihood-ratio test were used to identify the best-fit model associating age, sex, and presence of TILs with PD.
Results
Among 370 patients treated with CIT, 232 (62.7%) had dermatopathology reports for primary cutaneous melanoma; 142 reports (61.2%) classified TILs. Of 142 patients, 72 (50.7%) had PD while receiving CIT and 70 (49.3%) had NPD; 46 tumor specimens (32.4%) had absent TILs; 80 (56.3%), nonbrisk TILs; and 16 (11.3%), brisk TILs (Table). On univariate analysis, PD was not associated with age (odds ratio [OR], 1.01; 95% CI, 0.99-1.03; P = .27) or male sex (OR, 2.03; 95% CI, 0.98-4.20; P = .053).
Table. Demographic and Disease Characteristics of the 142 Patients Included in the Cohort With vs Without Disease Progression by TILs Classification.
| Characteristic | Patientsa | |||||
|---|---|---|---|---|---|---|
| Progression | No progression | Total | ||||
| Brisk or nonbrisk TILs | Absent TILs | Brisk or nonbrisk TILs | Absent TILs | |||
| Total | 42 (29.6) | 30 (21.1) | 54 (38.0) | 16 (11.2) | 142 (100) | |
| Age, mean (SD), y | 64.9 (14.6) | 60.8 (17.9) | 60.0 (15.3) | 61.2 (15.0) | NA | |
| Sex | ||||||
| Female | 7 (4.9) | 10 (7.0) | 21 (14.8) | 6 (4.2) | 44 (31.0) | |
| Male | 35 (24.6) | 20 (14.1) | 33 (23.2) | 10 (7.0) | 98 (69.0) | |
| Raceb | ||||||
| African American | 1 (0.7) | 0 | 0 | 0 | 1 (0.7) | |
| White | 37 (26.1) | 26 (18.3) | 46 (32.4) | 14 (9.9) | 123 (86.6) | |
| Other or unknown | 4 (2.8) | 4 (2.8) | 8 (5.6) | 2 (1.4) | 18 (12.7) | |
| Primary lesion location | ||||||
| Head and neck | 13 (9.2) | 10 (7.0) | 18 (12.7) | 6 (4.2) | 47 (33.1) | |
| Back and buttock | 8 (5.6) | 6 (4.2) | 13 (9.2) | 4 (2.8) | 31 (21.8) | |
| Chest and abdomen | 10 (7.0) | 2 (1.4) | 1 (0.7) | 0 | 13 (9.2) | |
| Upper extremity | 2 (1.4) | 5 (3.5) | 10 (7.0) | 2 (1.4) | 19 (13.4) | |
| Lower extremity | 9 (6.3) | 5 (3.5) | 12 (8.5) | 2 (1.4) | 28 (19.7) | |
| Anogenital | 0 | 2 (1.4) | 0 | 2 (1.4) | 4 (2.8) | |
| Therapy duration, mean (SD), d | 255.1 (370.2) | 193.2 (234.1) | 459.8 (371.0) | 403.8 (243.2) | NA | |
| Therapy type | ||||||
| Pembrolizumab | 39 (27.5) | 26 (18.3) | 53 (37.3) | 14 (9.9) | 132 (93.0) | |
| Nivolumab | 2 (1.4) | 3 (2.1) | 1 (0.7) | 1 (0.7) | 7 (4.9) | |
| Ipilimumab and nivolumab | 1 (0.7) | 1 (0.7) | 0 | 1 (0.7) | 3 (2.1) | |
Abbreviation: NA, not applicable; TILs, tumor-infiltrating lymphocytes.
Data are presented as number (percentage) of patients unless otherwise indicated.
Race was self-reported and was included to provide the cohort demographics. “Other or unknown” includes patients who declined to report or whose racial designation was not listed.
Patients with absent TILs had significantly increased rates of PD while receiving CIT vs those with present TILs (OR, 2.10; 95% CI, 1.07-4.11; P = .03) (Figure); 30 of 46 patients (65.2%) with absent, 34 of 80 (42.5%) with nonbrisk, and 8 of 16 (50.0%) with brisk TILs had PD. Compared with nonbrisk TILs, absence of TILs was associated with PD (OR, 2.90; 95% CI, 1.34-6.29; P = .01), but no significant difference was found between the brisk and absent TILs groups (OR, 1.88; 95% CI, 0.59-5.94; P = .29).
Figure. Rate of Disease Progression Among Patients Receiving Checkpoint Inhibitor Therapy by Tumor-Infiltrating Lymphocytes (TILs) Classification.
On multivariable analysis, a model for PD that included both sex (male: OR, 2.23; 95% CI, 1.05-4.73) and TILs (absent: OR, 2.60; 95% CI, 1.23-5.50) was a better fit than a model including only 1 of these variables by both AIC analysis and likelihood-ratio test. Age was not associated with PD when included in a multivariable model that contained sex and TILs.
Discussion
Overall, absence of primary TILs was associated with greater risk for PD among patients receiving CIT. That the brisk classification alone was not associated with decreased PD could be due to the small proportion with brisk TILs, which was a limitation. Alternatively, melanomas with brisk TILs may have developed variations and metastasized as less immunogenic clones. Multivariable analysis showed women had significantly lower rates of PD than men. Prior reports revealed conflicting data on whether men or women have improved survival associated with CIT.5,6
These results suggest that primary melanoma TILs classification could become a biomarker of metastatic disease response to CIT, and addition of sex may yield a better estimation model. Pathologists should consider including TILs classification in diagnostic reports if they do not already.
eMethods
eReference
References
- 1.Robert C. A decade of immune-checkpoint inhibitors in cancer therapy. Nat Commun. 2020;11(1):3801. doi: 10.1038/s41467-020-17670-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gibney GT, Weiner LM, Atkins MB. Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Lancet Oncol. 2016;17(12):e542-e551. doi: 10.1016/S1470-2045(16)30406-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Klein S, Mauch C, Brinker K, et al. Tumor infiltrating lymphocyte clusters are associated with response to immune checkpoint inhibition in BRAF V600E/K mutated malignant melanomas. Sci Rep. 2021;11(1):1834. doi: 10.1038/s41598-021-81330-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Straker RJ III, Krupp K, Sharon CE, et al. Prognostic significance of primary tumor-infiltrating lymphocytes in a contemporary melanoma cohort. Ann Surg Oncol. 2022;29(8):5207-5216. doi: 10.1245/s10434-022-11478-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Conforti F, Pala L, Bagnardi V, et al. Cancer immunotherapy efficacy and patients’ sex: a systematic review and meta-analysis. Lancet Oncol. 2018;19(6):737-746. doi: 10.1016/S1470-2045(18)30261-4 [DOI] [PubMed] [Google Scholar]
- 6.Jang SR, Nikita N, Banks J, et al. Association between sex and immune checkpoint inhibitor outcomes for patients with melanoma. JAMA Netw Open. 2021;4(12):e2136823. doi: 10.1001/jamanetworkopen.2021.36823 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods
eReference
