Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma

Michael R. Moore; Isabel D. Friesner; Emanuelle M. Rizk; Benjamin T. Fullerton; Manas Mondal; Megan H. Trager; Karen Mendelson; Ijeuru Chikeka; Tahsin Kurc; Rajarsi Gupta; Bethany R. Rohr; Eric J. Robinson; Balazs Acs; Rui Chang; Harriet Kluger; Bret Taback; Larisa J. Geskin; Basil Horst; Kevin Gardner; George Niedt; Julide T. Celebi; Robyn D. Gartrell-Corrado; Jane Messina; Tammie Ferringer; David L. Rimm; Joel Saltz; Jing Wang; Rami Vanguri; Yvonne M. Saenger

doi:10.1038/s41598-021-82305-1

Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma

Michael R. Moore, Isabel D. Friesner, Emanuelle M. Rizk, Benjamin T. Fullerton, Manas Mondal, Megan H. Trager, Karen Mendelson, Ijeuru Chikeka, Tahsin Kurc, Rajarsi Gupta, Bethany R. Rohr, Eric J. Robinson, Balazs Acs, Rui Chang, Harriet Kluger, Bret Taback, Larisa J. Geskin, Basil Horst, Kevin Gardner, George NiedtJulide T. Celebi, Robyn D. Gartrell-Corrado, Jane Messina, Tammie Ferringer, David L. Rimm, Joel Saltz, Jing Wang, Rami Vanguri, Yvonne M. Saenger

Research output: Contribution to journal › Article › peer-review

20 Scopus citations

Abstract

Accurate prognostic biomarkers in early-stage melanoma are urgently needed to stratify patients for clinical trials of adjuvant therapy. We applied a previously developed open source deep learning algorithm to detect tumor-infiltrating lymphocytes (TILs) in hematoxylin and eosin (H&E) images of early-stage melanomas. We tested whether automated digital (TIL) analysis (ADTA) improved accuracy of prediction of disease specific survival (DSS) based on current pathology standards. ADTA was applied to a training cohort (n = 80) and a cutoff value was defined based on a Receiver Operating Curve. ADTA was then applied to a validation cohort (n = 145) and the previously determined cutoff value was used to stratify high and low risk patients, as demonstrated by Kaplan–Meier analysis (p ≤ 0.001). Multivariable Cox proportional hazards analysis was performed using ADTA, depth, and ulceration as co-variables and showed that ADTA contributed to DSS prediction (HR: 4.18, CI 1.51–11.58, p = 0.006). ADTA provides an effective and attainable assessment of TILs and should be further evaluated in larger studies for inclusion in staging algorithms.

Original language	English (US)
Article number	2809
Journal	Scientific reports
Volume	11
Issue number	1
DOIs	https://doi.org/10.1038/s41598-021-82305-1
State	Published - Dec 2021
Externally published	Yes

ASJC Scopus subject areas

General

Access to Document

10.1038/s41598-021-82305-1

Cite this

Moore, M. R., Friesner, I. D., Rizk, E. M., Fullerton, B. T., Mondal, M., Trager, M. H., Mendelson, K., Chikeka, I., Kurc, T., Gupta, R., Rohr, B. R., Robinson, E. J., Acs, B., Chang, R., Kluger, H., Taback, B., Geskin, L. J., Horst, B., Gardner, K., ... Saenger, Y. M. (2021). Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma. Scientific reports, 11(1), Article 2809. https://doi.org/10.1038/s41598-021-82305-1

Moore, MR, Friesner, ID, Rizk, EM, Fullerton, BT, Mondal, M, Trager, MH, Mendelson, K, Chikeka, I, Kurc, T, Gupta, R, Rohr, BR, Robinson, EJ, Acs, B, Chang, R, Kluger, H, Taback, B, Geskin, LJ, Horst, B, Gardner, K, Niedt, G, Celebi, JT, Gartrell-Corrado, RD, Messina, J, Ferringer, T, Rimm, DL, Saltz, J, Wang, J, Vanguri, R & Saenger, YM 2021, 'Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma', Scientific reports, vol. 11, no. 1, 2809. https://doi.org/10.1038/s41598-021-82305-1

@article{09391547a88641edacc774bf3f324649,

title = "Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma",

abstract = "Accurate prognostic biomarkers in early-stage melanoma are urgently needed to stratify patients for clinical trials of adjuvant therapy. We applied a previously developed open source deep learning algorithm to detect tumor-infiltrating lymphocytes (TILs) in hematoxylin and eosin (H&E) images of early-stage melanomas. We tested whether automated digital (TIL) analysis (ADTA) improved accuracy of prediction of disease specific survival (DSS) based on current pathology standards. ADTA was applied to a training cohort (n = 80) and a cutoff value was defined based on a Receiver Operating Curve. ADTA was then applied to a validation cohort (n = 145) and the previously determined cutoff value was used to stratify high and low risk patients, as demonstrated by Kaplan–Meier analysis (p ≤ 0.001). Multivariable Cox proportional hazards analysis was performed using ADTA, depth, and ulceration as co-variables and showed that ADTA contributed to DSS prediction (HR: 4.18, CI 1.51–11.58, p = 0.006). ADTA provides an effective and attainable assessment of TILs and should be further evaluated in larger studies for inclusion in staging algorithms.",

author = "Moore, {Michael R.} and Friesner, {Isabel D.} and Rizk, {Emanuelle M.} and Fullerton, {Benjamin T.} and Manas Mondal and Trager, {Megan H.} and Karen Mendelson and Ijeuru Chikeka and Tahsin Kurc and Rajarsi Gupta and Rohr, {Bethany R.} and Robinson, {Eric J.} and Balazs Acs and Rui Chang and Harriet Kluger and Bret Taback and Geskin, {Larisa J.} and Basil Horst and Kevin Gardner and George Niedt and Celebi, {Julide T.} and Gartrell-Corrado, {Robyn D.} and Jane Messina and Tammie Ferringer and Rimm, {David L.} and Joel Saltz and Jing Wang and Rami Vanguri and Saenger, {Yvonne M.}",

note = "Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = dec,

doi = "10.1038/s41598-021-82305-1",

language = "English (US)",

volume = "11",

journal = "Scientific reports",

issn = "2045-2322",

publisher = "Nature Publishing Group",

number = "1",

}

TY - JOUR

T1 - Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma

AU - Moore, Michael R.

AU - Friesner, Isabel D.

AU - Rizk, Emanuelle M.

AU - Fullerton, Benjamin T.

AU - Mondal, Manas

AU - Trager, Megan H.

AU - Mendelson, Karen

AU - Chikeka, Ijeuru

AU - Kurc, Tahsin

AU - Gupta, Rajarsi

AU - Rohr, Bethany R.

AU - Robinson, Eric J.

AU - Acs, Balazs

AU - Chang, Rui

AU - Kluger, Harriet

AU - Taback, Bret

AU - Geskin, Larisa J.

AU - Horst, Basil

AU - Gardner, Kevin

AU - Niedt, George

AU - Celebi, Julide T.

AU - Gartrell-Corrado, Robyn D.

AU - Messina, Jane

AU - Ferringer, Tammie

AU - Rimm, David L.

AU - Saltz, Joel

AU - Wang, Jing

AU - Vanguri, Rami

AU - Saenger, Yvonne M.

PY - 2021/12

Y1 - 2021/12

N2 - Accurate prognostic biomarkers in early-stage melanoma are urgently needed to stratify patients for clinical trials of adjuvant therapy. We applied a previously developed open source deep learning algorithm to detect tumor-infiltrating lymphocytes (TILs) in hematoxylin and eosin (H&E) images of early-stage melanomas. We tested whether automated digital (TIL) analysis (ADTA) improved accuracy of prediction of disease specific survival (DSS) based on current pathology standards. ADTA was applied to a training cohort (n = 80) and a cutoff value was defined based on a Receiver Operating Curve. ADTA was then applied to a validation cohort (n = 145) and the previously determined cutoff value was used to stratify high and low risk patients, as demonstrated by Kaplan–Meier analysis (p ≤ 0.001). Multivariable Cox proportional hazards analysis was performed using ADTA, depth, and ulceration as co-variables and showed that ADTA contributed to DSS prediction (HR: 4.18, CI 1.51–11.58, p = 0.006). ADTA provides an effective and attainable assessment of TILs and should be further evaluated in larger studies for inclusion in staging algorithms.

AB - Accurate prognostic biomarkers in early-stage melanoma are urgently needed to stratify patients for clinical trials of adjuvant therapy. We applied a previously developed open source deep learning algorithm to detect tumor-infiltrating lymphocytes (TILs) in hematoxylin and eosin (H&E) images of early-stage melanomas. We tested whether automated digital (TIL) analysis (ADTA) improved accuracy of prediction of disease specific survival (DSS) based on current pathology standards. ADTA was applied to a training cohort (n = 80) and a cutoff value was defined based on a Receiver Operating Curve. ADTA was then applied to a validation cohort (n = 145) and the previously determined cutoff value was used to stratify high and low risk patients, as demonstrated by Kaplan–Meier analysis (p ≤ 0.001). Multivariable Cox proportional hazards analysis was performed using ADTA, depth, and ulceration as co-variables and showed that ADTA contributed to DSS prediction (HR: 4.18, CI 1.51–11.58, p = 0.006). ADTA provides an effective and attainable assessment of TILs and should be further evaluated in larger studies for inclusion in staging algorithms.

UR - http://www.scopus.com/inward/record.url?scp=85100250579&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85100250579&partnerID=8YFLogxK

U2 - 10.1038/s41598-021-82305-1

DO - 10.1038/s41598-021-82305-1

M3 - Article

C2 - 33531581

AN - SCOPUS:85100250579

SN - 2045-2322

VL - 11

JO - Scientific reports

JF - Scientific reports

IS - 1

M1 - 2809

ER -

Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this