EP710 - Association between endoscopic and radiologic features of near-complete response with local regrowth in rectal cancer after total neoadjuvant therapy

ePoster Presenter(s)

• 

  Christina I. Lee, MD

  Research Scholar
  Memorial Sloan Kettering
  NA
  New York, NY, United States

  Disclosure information not submitted.

Author(s)

• SL

  Sabrina T. Lin, Master of Science

  Biostatistician
  Memorial Sloan Kettering
  New York, New York, United States

  Disclosure information not submitted.
• 

  Despoina Kanata, MD (she/her/hers)

  Research fellow
  Memorial Sloan Kettetring Cancer Center
  Memorial Sloan Kettering Cancer Center
  New York, NY, United States

  All of the relevant financial relationships listed below have been mitigated.
  This individual has no financial relationships with ineligible companies.
• MA

  Mohammad Ali Abbass, MD MPH

  Assistant Attending
  Memorial Sloan Kettering Cancer Center
  New York, New York, United States

  Disclosure information not submitted.
• 

  Georgios Karagkounis, MD

  Assistant Attending
  Memorial Sloan Kettering Cancer Center
  Memorial Sloan Kettering
  New York, NY, United States

  All of the relevant financial relationships listed below have been mitigated.
  This individual has no financial relationships with ineligible companies.
• 

  Maria Widmar, MD, MPH (she/her/hers)

  Assistant Attending
  Memorial Sloan Kettering Cancer Center
  Weill Cornell Medical Center
  New York, New York, United States

  All of the relevant financial relationships listed below have been mitigated.
  This individual has no relevant relationships with ineligible companies.
• 

  Emmanouil Pappou, MD, PhD

  Assistant Attending
  Memorial Sloan Kettering Cancer Center
  New York, New York, United States

  Disclosure information not submitted.
• IW

  Iris H. Wei, MD

  Assistant Attending
  Memorial Sloan Kettering Cancer Center
  New York, New York, United States

  Disclosure information not submitted.
• 

  Garrett M. M. Nash, MD, MPH

  Vice Chair of Quality and Safety
  Memorial Sloan Kettering Cancer Center
  New York, New York, United States

  Disclosure information not submitted.
• 

  Philip B. B. Paty, MD

  Attending Surgeon
  Memorial Sloan Kettering Cancer Center
  New York, New York, United States

  Disclosure information not submitted.
• 

  Martin R. Weiser, MD

  Vice Chair for Faculty Affairs
  Memorial Sloan Kettering Cancer Center
  Cornell Medical College
  New York, New York, United States

  Disclosure information not submitted.
• LQ

  Li-Xuan Qin, n/a

  Attending Biostatistician
  Memorial Sloan Kettering Cancer Center
  New York, New York, United States

  Disclosure information not submitted.

Senior Author(s)

• 

  Julio Garcia Aguilar

  Disclosure information not submitted.

Purpose/Background: In the OPRA trial, rectal cancer response to total neoadjuvant therapy (TNT) was stratified into three groups defined by endoscopic and MRI criteria selected by a panel of experts. Patients with incomplete clinical response at restaging were offered total mesorectal excision. Patients with complete clinical response (cCR) or near complete clinical response (nCR) were offered watch-and-wait (WW). Unfortunately, 20% of patients with cCR and 50% with nCR of the WW cohort developed regrowth during surveillance. In this study, we investigated the association between endoscopic and MRI features used to define nCR and local regrowth in patients offered WW after restaging.

Methods/Interventions: Patients with stage II-III rectal adenocarcinoma treated with TNT - chemoradiation therapy (CRT) followed by consolidation chemotherapy or induction chemotherapy followed by CRT - were assessed for tumor response using endoscopy and MRI 8 ± 4 weeks after completion of treatment. The criteria for response are listed in Table 1. Patients who were offered WW based on tumor response at restaging were included in the analysis. Endoscopic and MRI features of nCR associated with increased risk of local regrowth within the WW cohort were assessed using univariable logistic regression model and multivariable backward selection model. A machine learning classification and regression tree (CART) model was then used to identify predictors of regrowth.

Results/Outcomes: Of the 324 patients randomized for the OPRA trial, 207 patients were offered WW and had follow-up of at least 2 years. The median follow-up for survivors was 4.9 years (IQR 4.1, 5.3). Median age was 59 (IQR 50, 68), 70 (34%) were female, and 110 (53%) were treated with consolidation TNT. Eighty-nine patients (43%) were categorized as endoscopic nCR and 94 patients (45.4%) as radiologic nCR at restaging. Regrowth occurred in 72 patients, while 135 patients had sustained cCR. Regrowth was associated with all MRI and endoscopic features of nCR except for erythema in univariate analysis. The presence of small nodules (OR 3.9, 95% CI 1.4 - 11.4; p=0.011) and superficial ulceration (OR 11.7, 95% CI 3.1 - 57.9; p< 0.001) on endoscopic exam, and partial regression or visible lymph nodes (OR 3.0, 95% CI 1.0 - 8.5; p=0.043) on rectal MRI were associated with increased risks of tumor regrowth in multivariable analysis. A CART model identified superficial ulceration as the only significant predictor for regrowth.

Conclusion/Discussion: While almost all endoscopic and radiological features used to define a nCR are individually associated with increased risks of tumor regrowth during WW, ulceration on endoscopy shows a stronger and independent association. This information should be used to simplify the classification of tumor response and improve the selection of rectal cancer patients for organ preservation.

Table 1. Summary of endoscopic and radiological features of response.