Mikroskoptan Yapay Zekâya: Küçük Hücreli Dışı Akciğer Kanserinde PD-L1 Testine Pratik Yaklaşım

Hatice Elmas , Burak Uzel , Abdullah Fahri ŞAHİN , Lutz WELKER

  • Burak Uzel: Virasoft Research and Artificial Intelligence Department, New York, USA
  • Abdullah Fahri ŞAHİN: School of Medicine, Malatya Turgut Ozal University
  • Lutz WELKER: University Medical Center Hamburg-Eppendorf (UKE)
  • Year : 2026
  • Vol : 42
  • Issue : 1
  •  Page : 71-79
Amaç: PD-L1 immünohistokimyası (IHC), küçük hücreli dışı akciğer kanseri (KHDAK) olan bireylerde immün kontrol noktası inhibitörü (ICI) tedavisini yönlendiren
temel bir prediktif biyobelirteç testidir. Ancak antikor klonlarındaki, derecelendirme sistemlerindeki (Tümör Oranı Skoru (TPS), Kombine Pozitif Skor (CPS), İmmün
Hücre skoru (IC)) ve pre-analitik/analitik koşullardaki değişkenlik, özellikle KHDAK'ta küçük biyopsiler ve sitolojik örneklerde yorumlamayı ve tekrarlanabilirliği
zorlaştırmaktadır. Bu derlemede, tümör heterojenliği, sitolojik kısıtlılıklar ve yapay zekâ (AI) tabanlı dijital patoloji araçlarının gelişen rolü vurgulanarak, KHDAK’ta
PD-L1 testindeki güncel uygulamaları, zorlukları ve gelişmeleri gözden geçirmek amaçlanmıştır. Ayrıca radyomiklerin de dahil olduğu multimodal yaklaşımların doku
temelli değerlendirmeyi nasıl tamamlayabileceğini ve ICI tedavisi için hasta seçiminde iyileşme sağlayıp sağlayamayacağını incelemeyi hedefledik.
Gereç ve Yöntemler: Ocak 2020 ile Haziran 2025 arasında yayımlanmış çalışmalar taranarak, KHDAK’ta PD-L1 ekspresyonunu doğrulanmış klonlar (22C3, 28-8, SP263,
SP142), sitoloji–histoloji uyumu, pre-analitik faktörler ve AI tabanlı PD-L1 skorlama platformları ile değerlendiren çalışmalar üzerine kapsamlı bir literatür taraması
yapıldı. Veriler, teknik değişkenlikleri, yorumlayıcı farklılıkları ve ortaya çıkan dijital çözümleri ele alan tematik bir sentez ile analiz edildi.
Bulgular: KHDAK’ta PD-L1 ekspresyonu, mekânsal heterojenite ve teknik değişkenliklerden etkilenmekte olup tanısal tutarsızlığa yol açmaktadır. Sitolojik örnekler,
sınırlı mimari yapı ve fiksasyon artefaktları nedeniyle benzersiz zorluklar oluşturur. Gözlemciler arası değişkenlik, özellikle %1–49 TPS aralığında en yüksektir. AI
destekli algoritmalar ve dijital platformlar, iyileştirilmiş tekrarlanabilirlik (κ 0.74’e kadar), doğruluk (%95’e kadar) ve klinik sonuçlarla olası korelasyon göstermiştir. Lunit
SCOPE PD-L1 ve HALO Lung PD-L1 AI gibi ticari AI platformları, %92’ye varan doğruluk oranı elde etmiş ve sınırda sonuçlardaki yanlış sınıflandırma oranlarını %18–30
azaltmıştır. PET tabanlı görüntüleme kullanan radyomikler—SUVmax, metabolik tümör hacmi ve heterojenlik indekslerini içeren—özellikle doku örneklemesinin sınırlı
olduğu durumlarda invaziv olmayan bir tamamlayıcı yöntem olarak umut vadetmektedir.
Sonuç: Güvenilir PD-L1 testi, klonlara özgü validasyon, standart protokollere uyum ve örnek sınırlılıklarının farkında olunmasını gerektirir. AI tabanlı dijital patoloji ve
radyomiklerin entegrasyonu, özellikle belirsiz veya sınırlı örneklerde tanısal doğruluğu artırabilir.
Cite this Article As : Elmas H, Uzel B, Şahin AF, Welker L. From Microscope to Machine: A Practical Guide to PD-L1 Testing in NSCLC. Selcuk Med J 2026;42(1): 71-79

Download Citation: Endnote/Zotero/Mendeley (RIS) RIS File

Download Citation: BibTeX BibTeX File

This article is published under the CC BY-NC 4.0 license.
Selcuk Medical Journal
2026, Vol. 42 (1)
ISSN: 1017-6616
E-ISSN: 2149-8059
Received : , Accepted : , Published Online :

References

  1. 1. Kawachi H, Yamada T, Tamiya M, et al. Clinical impact of cancer cachexia on the outcome of patients with non-small cell lung cancer with PD-L1 tumor proportion scores of ≥50% receiving pembrolizumab monotherapy versus immune checkpoint inhibitor with chemotherapy. Oncoimmunology. 2025;14(1):2442116.
  2. doi: 10.1080/2162402X.2024.2442116.
  3. 2. Haragan A, Field JK, Davies MPA, et al. Heterogeneity of PD-L1 expression in non-small cell lung cancer: Implications for specimen sampling in predicting treatment response. Lung Cancer. 2019;134:79–84. doi: 10.1016/j.lungcan.2019.06.005.
  4. 3. Kuempers C, van der Linde LIS, Reischl M, et al. Comparison of PD-L1 expression between paired cytologic and histologic specimens from non-small cell lung cancer patients. Virchows Arch. 2020;476(2):261–71. doi: 10.1007/s00428-019-02632-7.
  5. 4. Narote S, Desai SA, Patel VP, et al. Identification of new immune target and signaling for cancer immunotherapy. Cancer Genet. 2025;294–295:57–75.
  6. doi: 10.1016/j.cancergen.2025.03.004.
  7. 5. Carbone DP, Ciuleanu TE, Cobo M, et al. Nivolumab plus ipilimumab with chemotherapy as first-line treatment of patients with metastatic non-small-cell lung cancer: final, 6-year outcomes from CheckMate 9LA. ESMO Open. 2025;10(6):105123.
  8. doi: 10.1016/j.esmoop.2025.105123.
  9. 6. Paz-Ares L, Ciuleanu TE, Cobo M, et al. First-line nivolumab plus ipilimumab combined with two cycles of chemotherapy in patients with non-small-cell lung cancer (CheckMate 9LA): an international, randomised, open-label, phase 3 trial. Lancet Oncol. 2021;22(2):198–211.
  10. doi: 10.1016/S1470-2045(20)30641-0.
  11. 7. Zhou X, Xu Y, Ying Y, et al. Risk factors for checkpoint inhibitor pneumonitis in lung cancer patients treated with immune checkpoint inhibitors: a systematic review and meta-analysis. Front Immunol. 2025;16:1607170. doi: 10.3389/fimmu.2025.1607170.
  12. 8. Wiesweg M, Alaffas A, Rasokat A, et al. Treatment sequences in BRAF-V600-mutated NSCLC: first-line targeted therapy versus first-line (chemo-) immunotherapy. J Thorac Oncol. 2025;S1556-0864(25)00706-3. doi: 10.1016/j.jtho.2025.04.016.
  13. 9. Reck M, De T, Paz-Ares L, et al. Treatment-switching adjustment of overall survival in CheckMate 227 Part 1 evaluating first-line nivolumab plus ipilimumab versus chemotherapy for metastatic nonsmall cell lung cancer. Clin Lung Cancer. 2024;25(7):e362–8.
  14. doi: 10.1016/j.cllc.2024.06.005.
  15. 10. Plass M, Olteanu GE, Dacic S, et al. Comparative performance of PD-L1 scoring by pathologists and AI algorithms. Histopathology. 2025;87(1):90–100. doi: 10.1111/his.15432.
  16. 11. Cerbelli B, Girolami I, Eccher A, et al. Evaluating programmed death-ligand 1 (PD-L1) in head and neck squamous cell carcinoma: concordance between the 22C3 PharmDx assay and the SP263 assay on whole sections from a multicentre study. Histopathology. 2022;80(2):397–406. doi: 10.1111/his.14562.
  17. 12. Klempner SJ, Cowden ES, Cytryn SL, et al. PD-L1 immunohistochemistry in gastric cancer: comparison of combined positive score and tumor area positivity across 28-8, 22C3, and SP263 assays. JCO Precis Oncol. 2024;8:e2400230. doi: 10.1200/PO.24.00230.
  18. 13. Vranic S, Gatalica Z. PD-L1 testing by immunohistochemistry in immuno-oncology. Biomol Biomed. 2023;23(1):15–25. doi: 10.17305/bjbms.2022.7953.
  19. 14. Koomen BM, van der Starre-Gaal J, Vonk JM, et al. Formalin fixation for optimal concordance of programmed death-ligand 1 immunostaining between cytologic and histologic specimens from patients with non-small cell lung cancer. Cancer Cytopathol. 2021;129(4):304–17. doi: 10.1002/cncy.22383.
  20. 15. Elmas H, Diel R, Önal B, et al. Recommendations for immunocytochemistry in lung cancer typing: an update on a resource-efficient approach with large-scale comparative Bayesian analysis. Cytopathology. 2022;33(1):65–76. doi: 10.1111/cyt.13051.
  21. 16. Gosney JR, Boothman AM, Ratcliffe M, et al. Cytology for PD-L1 testing: a systematic review. Lung Cancer. 2020;141:101–6. doi: 10.1016/j.lungcan.2020.01.010.
  22. 17. Daverio M, Patrucco F, Gavelli F, et al. Comparative analysis of programmed death ligand 1 expression in paired cytologic and histologic specimens of non-small cell lung cancer. Cancer Cytopathol. 2020;128(8):580–8. doi: 10.1002/cncy.22292.
  23. 18. Paliogiannis P, Lobrano R, Bella MA, et al. PD-L1 immunohistochemical expression in bladder urothelial cancer with SP263, SP142 and 22C3 antibodies: a comparative study. Ann Diagn Pathol. 2024;69:152267. doi: 10.1016/j.anndiagpath.2024.152267.
  24. 19. Tejerina E, García Tobar L, Echeveste JI, et al. PD-L1 in cytological samples: a review and a practical approach. Front Med. 2021;8:668612. doi: 10.3389/fmed.2021.668612.
  25. 20. Lozano MD, Argueta A, de Andrea C. Immunotherapy and lung cytopathology: overview and possibilities. Cytopathology. 2024;35(2):213–7. doi: 10.1111/cyt.13335.
  26. 21. Zhang X, Zhang G, Qiu X, et al. Exploring non-invasive precision treatment in non-small cell lung cancer patients through deep learning radiomics across imaging features and molecular phenotypes. Biomark Res. 2024;12(1):12. doi: 10.1186/s40364-024-00561-5.
  27. 22. He W, Wu X, Jin Z, et al. Generative artificial intelligence in medical imaging: Current landscape, challenges, and future directions. Interdiscip Med. 2025;3(4).
  28. doi: 10.1002/INMD.20250024.
  29. 23. Elmas H, Uğuz AH, Şahin AF, Tecellioğlu FS, Welker L. Artificial intelligence, machine learning, and radiomics in lung cancer classification. Eur Res J. 2025:1–7.
  30. doi: 10.18621/eurj.1660161.
  31. 24. Bisson KR, Beharry A, Blais N, et al. Novel approach to proficiency testing reveals significant variations in biomarker practice leading to critical differences in lung cancer management. JTO Clin Res Rep. 2025;6(7):100837. doi: 10.1016/j.jtocrr.2025.100837.
  32. 25. Butter R, Halfwerk H, Radonic T, et al. The impact of impaired tissue fixation in resected non-small-cell lung cancer on protein deterioration and DNA degradation. Lung Cancer. 2023;178:108–115. doi: 10.1016/j.lungcan.2023.02.007.
  33. 26. Akhtar M, Rashid S, Al-Bozom IA. Correction: PD-L1 immunostaining: What pathologists need to know. Diagn Pathol. 2022;17(1):50. doi: 10.1186/s13000-022-01229-0.
  34. 27. Sholl LM, Awad M, Basu Roy U, Beasley MB, Cartun RW, Hwang DM, et al. Programmed death ligand-1 and tumor mutation burden testing of patients with lung cancer for selection of immune checkpoint inhibitor therapies: Guideline from the College of American Pathologists, Association for Molecular Pathology, International Association for the Study of Lung Cancer. Arch Pathol Lab Med. 2024;148(7):757–774. doi: 10.5858/arpa.2023-0379-SA.
  35. 28. Scheel AH, Baenfer G, Baretton G, Dietel M, Diezko R, Henkel T, et al. Interlaboratory concordance of PD-L1 immunohistochemistry for non-small-cell lung cancer. Histopathology. 2018;72(3):449–459. doi: 10.1111/his.13375.
  36. 29. Fang W, Jin H, Zhou H, Hong S, Ma Y, Zhang Y, et al. Intratumoral heterogeneity as a predictive biomarker in anti-PD-(L)1 therapies for non-small cell lung cancer. Molecular Cancer. 2021;20(1):37. doi: 10.1186/s12943-021-01331-9.
  37. 30. Büttner R, Gosney JR, Skov BG, Adam J, Motoi N, Bloom KJ, et al. Programmed death-ligand 1 immunohistochemistry testing: A review of analytical assays and clinical implementation in non–small-cell lung cancer. Journal of Clinical Oncology. 2017;35(34):3867–3876. doi: 10.1200/JCO.2017.74.7642.
  38. 31. Zdrenka M, Kowalewski A, Ahmadi N, Sadiqi RU, Chmura Ł, Borowczak J, et al. Refining PD-1/PD-L1 assessment for biomarker-guided immunotherapy: A review. Biomolecules and Biomedicine. 2024;24(1):14–29. doi: 10.17305/bjbms.2023.8368.
  39. 32. Jiao F, Shang Z, Lu H, Chen P, Chen S, Xiao J, et al. A weakly supervised deep learning framework for automated PD-L1 expression analysis in lung cancer. Front Immunol. 2025;16:1540087. doi: 10.3389/fimmu.2025.1540087.
  40. 33. Noll B, Wang W, Gong Y, Zhao J, Kalhor N, Prieto V, et al. Programmed death ligand 1 testing in non–small cell lung carcinoma cytology cell block and aspirate smear preparations. Cancer Cytopathol. 2018;126(5):342–52. doi: 10.1002/cncy.21987.
  41. 34. Jamal-Hanjani M, Wilson GA, McGranahan N, Birkbak NJ, Watkins TBK, Veeriah S, et al. Tracking the evolution of non–small-cell lung cancer. N Engl J Med. 2017;376(22):2109–21. doi: 10.1056/NEJMoa1616288.
  42. 35. Satturwar S, Girolami I, Munari E, Ciompi F, Eccher A, Pantanowitz L. Program death ligand‐1 immunocytochemistry in lung cancer cytological samples: A systematic review. Diagn Cytopathol. 2022;50(6):313–23. doi: 10.1002/dc.24955.
  43. 36. Hanna MG, Reuter VE, Samboy J, England C, Corsale L, Fine SW, et al. Implementation of digital pathology offers clinical and operational increase in efficiency and cost savings. Arch Pathol Lab Med. 2019;143(12):1545–55. doi: 10.5858/arpa.2018-0541-OA.
  44. 37. Jose AM, Karthik G, Shetty PK, Bhat V. Imaging primary cardiac angiosarcoma: Case-based illustrations with multimodality imaging and emphasis on the role of positron emission tomography-CT. BMJ Case Rep. 2024;17(6):e259022. doi: 10.1136/bcr-2023-259022.
  45. 38. Acanfora G, Iaccarino A, Cerbelli B, Di Cristofano C, Bellevicine C, Barberis M, et al. Interobserver agreement in PD-L1 evaluation on cytological samples—SAMPLING project: A multi‐institutional, international study. Cancer Cytopathol. 2025;133(3).
  46. doi: 10.1002/cncy.70003.
  47. 39. Kukuk SB, Shah AI, Çayır S, Solmaz-Yılmaz G, Eroğlu B, Erdem E, et al. Automated invasive area detection algorithm for breast cancer. European Congress on Digital Pathology (ECDP 2025); 2025 May.
  48. 40. Arbeiten M, Yu Z, Bai W. LWDDASGAN for WSIS. arXiv Preprint. 2019. arXiv:1906.11118. https://doi.org/10.48550/arXiv.1906.
  49. 41. Kabir S, Chowdhury MEH, Sarmun R, Vranić S, Al Saady RM, Rose I, et al. A novel deep learning framework for automatic scoring of PD-L1 expression in non-small cell lung cancer. Biomolecules and Biomedicine. 2025. Advance online publication. https://doi.org/10.17305/bjbms.2025.9505.
  50. 42. Kim H, Kim S, Choi S, Park C, Park S, Pereira S, et al. Clinical validation of artificial intelligence-powered PD-L1 tumor proportion score interpretation for immune checkpoint inhibitor response prediction in non-small cell lung cancer. JCO Precis Oncol. 2024;8:e2300556. https://doi.org/10.1200/PO.23.00556.
  51. 43. van Eekelen L, Spronck J, Looijen-Salamon M, Vos S, Munari E, Girolami I, et al. Comparing deep learning and pathologist quantification of cell-level PD-L1 expression in non-small cell lung cancer whole-slide images. Sci Rep. 2024;14(1):7136. https://doi.org/10.1038/s41598-024-57067-1.
  52. 44. Huang Z, Wang S, Zhou J, Chen H, Li Y. PD-L1 scoring models for non-small cell lung cancer in China: Current status, AI-assisted solutions and future perspectives. Thorac Cancer. 2025;16(7). https://doi.org/10.1111/1759-7714.70042.
  53. 45. Mansour MSI, Hejny K, Johansson F, Mufti J, Vidis A, Mager U, et al. Factors influencing concordance of PD-L1 expression between biopsies and cytological specimens in non-small cell lung cancer. Diagnostics. 2021;11(10):1927. https://doi.org/10.3390/diagnostics11101927.
  54. 46. Fujimoto D, Yamashita D, Fukuoka J, Kitamura Y, Hosoya K, Kawachi H, et al. Comparison of PD-L1 assays in non-small cell lung cancer: 22C3 pharmDx and SP263. Anticancer Res. 2018;38(12):6891–6895. https://doi.org/10.21873/anticanres.13065.
  55. 47. Sughayer MA, Alnaimy F, Alsughayer AM, Qamhia N. Comparison of 22C3 pharmDx and SP263 assays to test PD-L1 expression in NSCLC. Appl Immunohistochem Mol Morphol. 2019;27(9):663–6. doi:10.1097/PAI.0000000000000671.
  56. 48. Mansour MSI, Lindquist KE, Seidal T, Mager U, Mohlin R, Tran L, et al. PD-L1 testing in cytological non-small cell lung cancer specimens: A comparison with biopsies and review of the literature. Acta Cytol. 2021;65(6):501–9. doi:10.1159/000517078.
  57. 49. Rakaee M, Tafavvoghi M, Ricciuti B, Alessi JV, Cortellini A, Citarella F, et al. Deep learning model for predicting immunotherapy response in advanced non−small cell lung cancer. JAMA Oncol. 2025;11(2):109. doi:10.1001/jamaoncol.2024.6074.
  58. 50. Kim D, Sundling KE, Virk R, Thrall MJ, Alperstein S, Bui MM, et al. Digital cytology part 1: Digital cytology implementation for practice: A concept paper with review and recommendations from the American Society of Cytopathology Digital Cytology Task Force. J Am Soc Cytopathol. 2024;13(2):86–96. doi:10.1016/j.jasc.2023.10.007.
  59. 51. Kholová I. Cyto-histopathological correlations in pathology diagnostics. Diagnostics. 2022;12(7):1240. doi:10.3390/diagnostics12071240.
  60. 52. Chantziantoniou N, Mukherjee M, Donnelly AD, Pantanowitz L, Austin RM. Digital applications in cytopathology: Problems, rationalizations, and alternative approaches. Acta Cytol. 2018;62(1):68–76. doi:10.1159/000484434.
  61. 53. Evans AJ, Brown RW, Bui MM, Chlipala EA, Lacchetti C, Milner DA, et al. Validating whole slide imaging systems for diagnostic purposes in pathology. Arch Pathol Lab Med. 2022;146(4):440–450. doi:10.5858/arpa.2020-0723-RA.
  62. 54. Kim D, Sundling KE, Virk R, Thrall MJ, Alperstein S, Bui MM, et al. Digital cytology part 2: Artificial intelligence in cytology: A concept paper with review and recommendations from the American Society of Cytopathology Digital Cytology Task Force. J Am Soc Cytopathol. 2024;13(2):97–110. doi:10.1016/j.jasc.2023.10.006.
  63. 55. Sakamoto T, Furukawa T, Lami K, Pham HHN, Uegami W, Kuroda K, et al. A narrative review of digital pathology and artificial intelligence: Focusing on lung cancer. Transl Lung Cancer Res. 2020;9(5):2255–2276. doi:10.21037/tlcr-20-616.
  64. 56. Binczyk F, Prazuch W, Bozek P, Polanska J. Radiomics and artificial intelligence in lung cancer screening. Transl Lung Cancer Res. 2021;10(2):1186–1199. doi:10.21037/tlcr-20-838.
  65. 57. Quiros CA, Coudray N, Yeaton A, Yang X, Liu B, Le H, et al. Mapping the landscape of histomorphological cancer phenotypes using self-supervised learning on unannotated pathology slides. Nat Commun. 2024;15(1):4596. doi:10.1038/s41467-024-48666-7.
  66. 58. Hanna MG, Ardon O, Reuter VE, Sirintrapun SJ, England C, Klimstra DS, et al. Integrating digital pathology into clinical practice. Mod Pathol. 2022;35(2):152–164. doi:10.1038/s41379-021-00922-0.
  67. 59. Caiado F, Silva‐Santos B, Norell H. Intra‐tumour heterogeneity – going beyond genetics. FEBS J. 2016;283(12):2245–2258. doi:10.1111/febs.13705.
  68. 60. Abdo M, Belloum Y, Heigener D, Welker L, von Weihe S, Schmidt M, et al. Comparative evaluation of PD‐L1 expression in cytology imprints, circulating tumour cells and tumour tissue in non‐small cell lung cancer patients. Mol Oncol. 2023;17(5):737–746. doi:10.1002/1878-0261.13415.