Pathologic Complete Response In Patients Research Articles

OT2-05-02: ACRIN 6691 Monitoring and Predicting Breast Cancer Neoadjuvant Chemotherapy Response Using Diffuse Optical Spectroscopic Imaging (DOSI).

Abstract Background Imaging technologies monitoring and predicting breast cancer response to neoadjuvant chemotherapy (NAC) are of increasing interest. The utility of conventional imaging approaches varies and identifies the need for alternate functional imaging strategies. The use of model-based photon migration methods to quantitatively separate light absorption from scattering in multiply-scattering tissues is a type of near-infrared spectroscopy (NIRS) broadly referred to as diffuse optical spectroscopy (DOS) [Bevilacqua, et al. Applied Optics, 2000; Jakubowski, et al., J of Applied Optics, 2009]. DOSI is a promising experimental technology that allows patients undergoing NAC to be followed with a “no significant risk” device meeting Food and Drug Administration criteria for exempt status. The current design is a mobile device which offers increased accessibility and is relatively simple to perform and interpret, as compared to mammography, magnetic resonance imaging, and positron emission tomography. Due to its size and portability, DOSI is a low barrier-to-access technology, creating new opportunities for patients to receive personalized treatment and for physicians to gain new insight into response mechanisms. The long-term goal is to provide oncologists with a relatively simple, risk-free bedside tool that can be used to help inform medical decisions on chemotherapy regimen, duration, and timing of surgery, thereby maximizing therapeutic response and minimizing unnecessary toxicity. Trial design: In this phase I/II prospective single arm study, patients will receive SOC NAC at five (5) NCI Network for Translational Research in Optical Imaging (NTROI) clinical sites with identical DOSI instruments and procedures. Patients will receive four DOSI exams: at baseline before chemotherapy, at early therapy 5–10 days after NAC initiation, at mid therapy, and at post therapy prior to surgery. The protocol will evaluate a harmonized DOSI technology platform that has been standardized for NAC monitoring. Eligibility: Women who have been diagnosed with breast cancer, have had confirmation by pre-treatment biopsy, and are scheduled to receive NAC followed by surgery are eligible for this trial. Specific aims: The primary aim of this clinical trial is to determine whether the baseline to mid-therapy changes in the DOSI measurement of the quantitative tumor tissue optical index can predict final pathologic complete response in patients with breast cancer undergoing NAC. The secondary aims investigate the correlation between additional DOSI quantitative measurements of tumor biochemical composition obtained at other timepoints, the full range of pathologic response (i.e. complete, partial, and non-response), and any corresponding imaging measurements. Statistical methods: Logistic regression models will be used to study the relationships between pathological complete response and percent change in tissue optical index tumor to normal ratio at different imaging time points. Study size: A total of sixty (60) patients will be enrolled in this imaging study. Currently, one patient has accrued. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr OT2-05-02.

Accuracy of Clinical Examination, Digital Mammogram, Ultrasound, and MRI in Determining Postneoadjuvant Pathologic Tumor Response in Operable Breast Cancer Patients

To determine the accuracy, positive predictive value (PPV), and negative predictive value (NPV) of clinical examination and breast imaging techniques in determining pathologic complete response in patients with locally advanced breast cancer after neoadjuvant therapy. A retrospective review was performed of data collected from patients treated with either neoadjuvant hormonal or chemotherapy between January 2005 and September 2010. Patients were evaluated by one of three surgical breast oncologists before neoadjuvant therapy and within 1 month before surgery by clinical breast examination (CBE), digital mammogram, breast ultrasound, and/or magnetic resonance imaging (MRI). The accuracy, NPV, and PPV of each modality was calculated on the basis of the final pathologic report. Available data from the literature was synthesized. Sixty-two tumors in 61 patients with a mean age of 56 (range 34-87) years were evaluated. Overall accuracy ranged from 54% (CBE) to 80% (breast ultrasound). All modalities had a PPV greater than 75% for identifying the presence of residual disease. The PPV of each modality was generally higher in the younger patients. The NPV of all methods was less than 50%. The accuracy and NPV were compromised even further in younger patients. The combination of our own data with data available from the literature revealed MRI to be superior with regard to accuracy and PPV, but the NPV of MRIs remained poor at 65%. All measured tests are good at predicting the presence of disease on final pathology, but none are able to reliably predict a pathologic complete response.

Functional proteomics can define prognosis and predict pathologic complete response in patients with breast cancer.

PurposeTo determine whether functional proteomics improves breast cancer classification and prognostication and can predict pathological complete response (pCR) in patients receiving neoadjuvant taxane and anthracycline-taxane-based systemic therapy (NST).MethodsReverse phase protein array (RPPA) using 146 antibodies to proteins relevant to breast cancer was applied to three independent tumor sets. Supervised clustering to identify subgroups and prognosis in surgical excision specimens from a training set (n = 712) was validated on a test set (n = 168) in two cohorts of patients with primary breast cancer. A score was constructed using ordinal logistic regression to quantify the probability of recurrence in the training set and tested in the test set. The score was then evaluated on 132 FNA biopsies of patients treated with NST to determine ability to predict pCR.ResultsSix breast cancer subgroups were identified by a 10-protein biomarker panel in the 712 tumor training set. They were associated with different recurrence-free survival (RFS) (log-rank p = 8.8 E-10). The structure and ability of the six subgroups to predict RFS was confirmed in the test set (log-rank p = 0.0013). A prognosis score constructed using the 10 proteins in the training set was associated with RFS in both training and test sets (p = 3.2E-13, for test set). There was a significant association between the prognostic score and likelihood of pCR to NST in the FNA set (p = 0.0021).ConclusionWe developed a 10-protein biomarker panel that classifies breast cancer into prognostic groups that may have potential utility in the management of patients who receive anthracycline-taxane-based NST.

Neoadjuvant Dose-Dense Gemcitabine plus Docetaxel and Vinorelbine plus Epirubicin for Operable Breast Cancer

Background: Neoadjuvant anti-tumor activity of an alternating taxane- and anthracycline-based dose-dense regimen in patients with operable, noninflammatory large breast cancer was investigated. Objective: The objective is to study the rate of pathological complete response in patients with breast cancer receiving dose-dense chemotherapy sequentially with gemcitabine plus docetaxel and vinorelbine plus epirubicin. Methods: Women (n = 74) with clinical stage II or III breast cancer were enrolled in this open-label, multicenter study to receive six 2-weekly courses of gemcitabine 1000 mg/m2 plus docetaxel 75 mg/m2 on days 1 and 15, and vinorelbine 25 mg/m2 plus epirubicin 100mg/m2 on days 29 and 43. Patients with an objective response on day 56 then received another cycle of gemcitabine/ docetaxel on day 57 and of vinorelbine/epirubicin on day 71. Conservative surgery was scheduled for all patients. Results: Of the patients enrolled, 30% had triple-negative breast cancer (TNBC). The pathologic complete response (pCR) rate was 22% overall, but was higher in TNBC than patients without TNBC (40.9% vs 14.0%; p=0.028). Among patients with a pCR, patients with TNBC had similar recurrence-free survival (RFS) and overall survival (OS) to patients without TNBC. Among those without a pCR, RFS rates for patients with TNBC were significantly lower than for patients without TNBC (p=0.04). The most common severe hematologic toxicity was neutropenia. Conclusions: Administering four drugs in a dose-dense alternating sequence gave a high pCR in patients with operable, invasive breast cancer. Patients with TNBC with a pCR had similar OS to patients without TNBC, whereas patients with TNBC without a pCR had poorer survival rate than their non- TNBC counterparts.

Neoadjuvant Dose-Dense Gemcitabine plus Docetaxel and Vinorelbine plus Epirubicin for Operable Breast Cancer

Background: Neoadjuvant anti-tumor activity of an alternating taxane- and anthracycline-based dose-dense regimen in patients with operable, noninflammatory large breast cancer was investigated.Objective: The objective is to study the rate of pathological complete response in patients with breast cancer receiving dose-dense chemotherapy sequentially with gemcitabine plus docetaxel and vinorelbine plus epirubicin.Methods: Women (n = 74) with clinical stage II or III breast cancer were enrolled in this open-label, multicenter study to receive six 2-weekly courses of gemcitabine 1000 mg/m2 plus docetaxel 75 mg/m2 on days 1 and 15, and vinorelbine 25 mg/m2 plus epirubicin 100mg/m2 on days 29 and 43. Patients with an objective response on day 56 then received another cycle of gemcitabine/ docetaxel on day 57 and of vinorelbine/epirubicin on day 71. Conservative surgery was scheduled for all patients.Results: Of the patients enrolled, 30% had triple-negative breast cancer (TNBC). The pathologic complete response (pCR) rate was 22% overall, but was higher in TNBC than patients without TNBC (40.9% vs 14.0%; p=0.028). Among patients with a pCR, patients with TNBC had similar recurrence-free survival (RFS) and overall survival (OS) to patients without TNBC. Among those without a pCR, RFS rates for patients with TNBC were significantly lower than for patients without TNBC (p=0.04). The most common severe hematologic toxicity was neutropenia.Conclusions: Administering four drugs in a dose-dense alternating sequence gave a high pCR in patients with operable, invasive breast cancer. Patients with TNBC with a pCR had similar OS to patients without TNBC, whereas patients with TNBC without a pCR had poorer survival rate than their non- TNBC counterparts.

ACRIN 6691 monitoring and predicting breast cancer neoadjuvant chemotherapy response using diffuse optical spectroscopic imaging (DOSI).

TPS249 Background: Imaging technologies monitoring and predicting breast cancer response to neoadjuvant chemotherapy (NAC) are of increasing interest. The utility of conventional imaging approaches varies and identifies the need for alternate functional imaging strategies. DOSI is a promising experimental technology that allows patients undergoing NAC to be followed with a “no significant risk” device meeting Food and Drug Administration criteria for exempt status. The current design is a mobile device which offers increased accessibility. DOSI is relatively simple to perform and interpret, as compared to mammography, magnetic resonance imaging, and positron emission tomography. Due to its size and portability, DOSI is a low barrier-to-access technology, creating new opportunities for patients to receive personalized treatment and for physicians to gain new insight into response mechanisms. Our long-term goal is to provide oncologists with a relatively simple, risk-free bedside tool that can be used to help inform medical decisions on chemotherapy regimen, duration, and timing of surgery, thereby maximizing therapeutic response and minimizing unnecessary toxicity. Methods: In this phase I/II multicenter study, women who have been diagnosed with breast cancer, have had confirmation by pre-treatment biopsy, and are scheduled to receive NAC followed by surgery are eligible for this trial. Approximately 60 patients receiving NAC will participate at five clinical sites. This protocol will evaluate a harmonized DOSI technology platform that has been standardized for NAC monitoring. The primary aim of this clinical trial is to determine whether the baseline to mid-therapy changes in the DOSI measurement of the quantitative tumor tissue optical index can predict final pathologic complete response in patients with breast cancer undergoing NAC. The secondary aims investigate the correlation between additional DOSI quantitative measurements of tumor biochemical composition obtained at other timepoints, the full range of pathologic response (i.e. complete, partial, and non-response), and any corresponding imaging measurements.

Abstract P2-06-12: HER2 mRNA Expression Is Influenced by Estrogen Receptor Status and PI3K Mutational Status

Abstract Molecular profiling has suggested that ER+/HER2+ breast cancer is a distinct entity from ER-/HER2+ breast cancer and confers a worse prognosis in systemically untreated patients. It has recently been reported that genes associated with PI3K/AKT pathway activation predict pathologic complete response in patients receiving trastuzumab-chemotherapy in ER+/HER+ but not in ER-/HER2+ tumors. Past studies have indicated that 30-40% of breast cancers harbor a PI3K mutation, predominantly in exons 9 and 20. We analyzed 100 HER2+ breast cancers stratified by ER status to confirm the differing level of HER2 mRNA expression relative to ER status and to determine the association, in any, to the presence of a PI3K mutation. Methods: Fifty ER+ and 50 ER-breast cancers previously assessed as HER2+ (HER2:CEP-17 ratio ranging from 2.40-16.20, PathVysion Kit, Abbott) were pulled from the archives. ER was assessed by IHC, (6F11, NovaCastra) using a clinically validated method in accordance with the new CAP/ASCO guidelines. Molecular analysis was performed on paraffin embedded tissues. Tumor areas of interest were identified and selectively removed from the section. RNA and DNA were isolated from the collected tissue. Real-time quantitative RT-PCR was used to evaluate the expression of the HER2 gene from the RNA fraction. The expression values were calculated as a ratio of HER2 to the reference gene. The DNA was added to an allele specific PCR to assess 7 of the most common mutations in exons 9 and 20 of the PI3K gene. All data was generated on an AB 7900 HT real-time platform. Results: Quantitative HER2 mRNA Results Discusssion: The level of HER2 mRNA expression differs significantly based on the expression of ER. This is further accentuated by the presence of a PI3K mutation, increasing the level of HER2 mRNA expression is ER-patients and decreasing the level of HER2 mRNA expression in ER+ patients. This observation needs to be further explored and may have profound implications, if as current data suggests, the subset of patients most responsive to trastuzumab therapy are the ER+, PI3K activated tumors. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-06-12.

Bevacizumab combined with chemotherapy significantly improves pathologic complete response in patients with operable or locally advanced breast cancer.

Abstract Abstract #5114 Bevacizumab (B) improved progression free survival in metastatic breast cancer (BC) when added to chemotherapy (CT) and was relatively safe in the adjuvant setting. This study evaluated pathologic complete response (pCR) and safety of B with CT in the neoadjuvant setting.
 Methods: This is a single arm, single institution phase II trial. Patients (Pts) eligible for the study were women with proven BC (T1-4 and/or positive axillary LN -excluding inflammatory BC). Pts received CT with docetaxel (T, 75 mg/m2), cyclophosphamide (C, 500 mg/m2) and B (15 mg/kg), TCB every 3 wks x 4 cycles followed by doxorubicin (A, 60 mg/m2) every 3 wks for 4 cycles. After CT, Pts were evaluated clinically (clinical complete or partial response cCR or cPR), by mammogram and for their heart function by MUGA. Breast conserving surgery was considered if the response was deemed satisfactory, otherwise mastectomy was performed and pCR was evaluated. 28 to 84 days after surgery and after documented healing of the operative incision, B was restarted at 15 mg/kg x 9 cycles. XRT, herceptin and endocrine therapy were given as indicated concurrently with B.
 Results: As planned, 40 Pts were enrolled. Average age 46 (range 27-73). One patient withdrew consent after 2 cycles of TCB. Average tumor size was 5.26 cm (range 1.8 - 15 cm). Lobular 7/ductal 32. Grade I/II/III: 3/12/24. ER +/PR +: 11; ER+/PR-8; ER-/ PR +: 3, ER-/PR-:17. HER2 +: 8. Triple negative: 14. Lymph nodes were + in 25 patients. 32 Pts completed all pre-op CT cycles. One Pt died after 2 cycles of TCB of bilateral PE, 4 Pts received a truncated doxorubicin course due to side effects, 2 Pt received TCB x 2 followed by A x 4. The most common grade 3/4 AEs (#): diarrhea (4), febrile neutropenia (4), musculoskeletal pain (10), fatigue (8), mucositis (3), nausea (2), hypertension (5), syncope (2) infection (3). No clinically significant change in LVEF was noticed (Pre-CT EF 63.3% (55-76%); post-CT EF 61.2% (49-72%)). 4 Pts had a decrease of EF > 10. 21 cCR and 16 cPR were seen.1 Pt had progression and 1 Pt had no change. 38 Pts underwent surgery. 10 had lumpectomy (average pre-CT size 4.8 cm) and 28 had mastectomy (average pre-CT size 6.1 cm). One Pt had infusaport infection and another had breast implant infection, both required device removal and had delayed wound healing. 16 pCR/39 in the breast and 13 pCR/39 in the breast and axilla were seen (pCR breast 41%, pCR breast and axilla 33%). Several Pts had near pCR. Post-operative B safety data will be presented.
 Conclusions: When added to CT, B improved pCR in a poor prognostic BC Pts and was associated with manageable toxicity. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 5114.

Positron Emission Tomography Scanning and Pathologic Assessment of Neoadjuvant Chemoradiotherapy for Esophageal Cancer

Positron Emission Tomography Scanning and Pathologic Assessment of Neoadjuvant Chemoradiotherapy for Esophageal Cancer

Predictive Factors for a Pathologic Complete Response in Patients Receiving Preoperative Chemoradiation for Locally Advanced Rectal Cancer

Preoperative chemoradiation (CRT) is the standard of care for patients with locally advanced rectal cancer. The factors that predict for a pathologic complete response (pCR), however, have yet to be fully defined. The purpose of this study, therefore, is to identify patient, tumor, and treatment factors that predict for a pCR in these patients. We retrospectively analyzed the treatment outcome for 53 adult patients with T3 and/or N+ rectal cancer who underwent preoperative CRT at our institution. Three patients (6%) had T2 N+ disease, 24 patients (45%) had T3 N0 disease, and 26 patients (49%) had T3 N+ disease. Radiation therapy consisted of 4,500 cGy delivered to the whole pelvis followed by a 540 cGy boost to the primary tumor and/or involved lymph nodes. Chemotherapy consisted of a fluoropyrimidine alone in 26 patients or a fluoropyrimidine with oxaliplatin in 27 patients. Serum carcinoembryonic antigen (CEA) was recorded at the time of diagnosis for 43 patients. The interval from the last day of CRT to the day of surgery was recorded for all patients. The pCR rate was 23% for our cohort of patients. On univariate analysis, 100% (3/3) of patients with T2 tumors had a CR compared with 18% (9/50) of patients with T3 tumors (p = 0.009 by Fisher's exact test). The interval between the end of CRT and surgery was of predictive significance (p = 0.02 by t test), in that patients who achieved a pCR had a mean interval of 72 days between the end of CRT and surgery and patients not achieving a pCR had a mean interval of 51 days. There was a trend (p = 0.17 by t test) for lower serum CEA at diagnosis to be associated with a pCR. Lymph node status, location of tumor within the rectum, patient age, patient gender, and the type of chemotherapy did not have an effect on pCR. On multivariate analysis, a longer time interval between CRT and surgery was associated with a 1.04-fold greater likelihood of a pCR (p = 0.02) and an increasing CEA at the time of diagnosis, when measured as a continuous variable, was associated with a 0.83-fold reduction in the likelihood of a pCR (p = 0.09). Our study suggests that a longer time interval between the end of CRT and surgery as well as a lower serum CEA at diagnosis are associated with higher pCR rates. Given the small number of patients in our study with T2 tumors, we can only infer that a lower T stage may also be associated with a higher pCR rate. Consequently, the optimal interval between CRT and surgery may be longer than the currently recommended 4 to 6 weeks and more aggressive preoperative therapy may be appropriate for patients with significantly elevated serum CEA levels at diagnosis. We believe that additional studies with larger sample sizes are warranted to confirm these findings.

Retraction—validation of gene signatures that predict the response of breast cancer to neoadjuvant chemotherapy: a substudy of the EORTC 10994/BIG 00-01 clinical trial

Summary Background We have previously described gene-expression signatures that predict growth inhibitory and cytotoxic effects of common chemotherapeutic drugs in vitro. The aim of this study was to confirm the validity of these gene-expression signatures in a large series of patients with oestrogen-receptor-negative breast tumours who were treated in a phase III neoadjuvant clinical trial. Methods This trial compares a non-taxane regimen (fluorouracil, epirubicin, and cyclophosphamide [FEC] for six cycles) with a taxane regimen (docetaxel for three cycles followed by epirubicin plus docetaxel [TET] for three cycles) in women with oestrogen-receptor-negative breast cancer. The primary endpoint of the study is the difference in progression-free survival based on TP53 status and will be reported later. Predicting response with gene signatures was a planned secondary endpoint of the trial and is reported here. Pathological complete response, defined as complete disappearance of the tumour with no more than a few scattered tumour cells detected by the pathologist in the resection specimen, was used to assess chemosensitivity. RNA was prepared from sections of frozen biopsies taken at diagnosis and hybridised to Affymetrix X3P microarrays. In-vitro single-agent drug sensitivity signatures were combined to obtain FEC and TET regimen-specific signatures. This study is registered on the clinical trials site of the US National Cancer Institute website http://www.clinicaltrials.gov/ct/show/NCT00017095. Findings Of 212 patients with oestrogen-receptor-negative tumours assessed, 87 patients were excluded. 125 oestrogen-receptor-negative tumours (55 that showed pathological complete responses) were tested: 66 in the FEC group (28 that showed pathological complete responses) and 59 in the TET group (27 that showed pathological complete responses). The regimen-specific signatures significantly predicted pathological complete response in patients treated with the appropriate regimen (p Interpretation We have validated the use of regimen-specific drug sensitivity signatures in the context of a multicentre randomised trial. The high NPV of both signatures may allow early selection of patients with breast cancer who should be considered for trials with new drugs.

117 Negative post-treatment 18-FDG pet scans may not predict pathologic complete response in patients with rectal cancer undergoing neoadjuvant chemoradiation

117 Negative post-treatment 18-FDG pet scans may not predict pathologic complete response in patients with rectal cancer undergoing neoadjuvant chemoradiation

Positive Emission Tomography for Evaluating a Complete Clinical Response in Patients with Ovarian or Peritoneal Carcinoma: Correlation with Second-Look Laparotomy

Objective. Positive emission tomography (PET) provides a novel means of imaging malignancies. The following study was undertaken to evaluate the predictive value of PET in determining a pathologic complete response in patients with advanced ovarian or peritoneal carcinoma who had a complete clinical response following primary chemotherapy.Methods. Twenty-two patients with advanced-stage ovarian (N = 17) or peritoneal (N = 5) carcinoma who had achieved complete clinical and radiologic remission and normal CA-125 level after six cycles of chemotherapy and who had consented to a second look laparotomy procedure were studied. All patients received platinum based therapy and all but one patient, treated elsewhere, received paclitaxel in combination with platinum. Following IV administration of 20 mCi [18F]fluorodeoxyglucose (FDG), the entire abdomen and pelvis were scanned. Various technical modifications including bladder activity dilution, intravenous hydration with diuretic therapy, and mechanical bowel preparations, were used to reduce background activity. Second-look laparotomy findings were classified as negative, macroscopically positive if a biopsy of a suspicious area was histologically positive, or microscopically positive if only a nonsuspicious area was histologically positive. The effect of patient preparation prior to PET imaging was evaluated.Results. Persistent disease was found in 13 of the 22 patients (59%). Only one of nine sites with macroscopic and none of four with microscopic disease were accurately predicted. The sensitivity was only 10% and the specificity 42%. Intravenous hydration, diuretic therapy, and bowel preparation did not improve the results.Conclusions. These results suggest that despite technical modifications the sensitivity of PET before second-look laparotomy for small-volume persistent disease is low.

Pathologic complete response in advanced non-small-cell lung cancer following preoperative chemotherapy: implications for the design of future non-small-cell lung cancer combined modality trials.

This report determines the incidence of pathologic complete response in patients with locally advanced non-small-cell lung cancer (NSCLC) treated with mitomycin, vinca alkaloid, and high-dose cisplatin (MVP) chemotherapy, and estimates the effect of MVP on survival. We have identified and reviewed the course of 21 patients with advanced NSCLC who achieved a pathologic complete response following a median of three preoperative MVP combination chemotherapy courses including vinblastine or vindesine, cisplatin (120 mg/m2), and mitomycin (n = 19). All patients had a major objective response following preoperative chemotherapy and nine (43%) had a clinical complete response. Nine patients with pathologic complete responses were among 73 entered on a preoperative chemotherapy program, yielding an incidence estimate of 12% (95% confidence interval, 6% to 22%). The median survival duration for all 21 patients has not been reached. The median follow-up duration is now 68 months (range, 17 to 109). Survival estimates are 90% at 1 year, 62% at 3 years, and 54% at 5 years. Nine patients have relapsed with initial sites of recurrence as follows: brain (n = 5), other systemic sites (n = 3), and locoregional (n = 1). One patient died in the postoperative period. Eleven patients remain disease-free and all have excellent functional status. We have observed pathologic complete responses in approximately 12% of advanced NSCLC patients treated with preoperative MVP chemotherapy. These pathologically determined responses were seen only in patients with major objective responses clinically. Pathologic complete response predicts excellent survival and functional level and should be considered a major end point in the evaluation of preoperative chemotherapy programs.

Study of the decomposition of TiCWC solid solutions: S Zhenhui, (Zhuzhou Cemented Carbide Industry Co, Hunan, China), Int J. Refractory Metals & Hard Materials, Vol 9, No 3, 1990, 154–156

Accurate prediction of tumour response to neoadjuvant chemoradiotherapy enables personalised perioperative therapy for locally advanced rectal cancer. We aimed to develop and validate an artificial intelligence radiopathomics integrated model to predict pathological complete response in patients with locally advanced rectal cancer using pretreatment MRI and haematoxylin and eosin (H&E)-stained biopsy slides.In this multicentre observational study, eligible participants who had undergone neoadjuvant chemoradiotherapy followed by radical surgery were recruited, with their pretreatment pelvic MRI (T2-weighted imaging, contrast-enhanced T1-weighted imaging, and diffusion-weighted imaging) and whole slide images of H&E-stained biopsy sections collected for annotation and feature extraction. The RAdioPathomics Integrated preDiction System (RAPIDS) was constructed by machine learning on the basis of three feature sets associated with pathological complete response: radiomics MRI features, pathomics nucleus features, and pathomics microenvironment features from a retrospective training cohort. The accuracy of RAPIDS for the prediction of pathological complete response in locally advanced rectal cancer was verified in two retrospective external validation cohorts and further validated in a multicentre, prospective observational study (ClinicalTrials.gov, NCT04271657). Model performances were evaluated using area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).Between Sept 25, 2009, and Nov 3, 2017, 303 patients were retrospectively recruited in the training cohort, 480 in validation cohort 1, and 150 in validation cohort 2; 100 eligible patients were enrolled in the prospective study between Jan 10 and June 10, 2020. RAPIDS had favourable accuracy for the prediction of pathological complete response in the training cohort (AUC 0·868 [95% CI 0·825–0·912]), and in validation cohort 1 (0·860 [0·828–0·892]) and validation cohort 2 (0·872 [0·810–0·934]). In the prospective validation study, RAPIDS had an AUC of 0·812 (95% CI 0·717–0·907), sensitivity of 0·888 (0·728–0·999), specificity of 0·740 (0·593–0·886), NPV of 0·929 (0·862–0·995), and PPV of 0·512 (0·313–0·710). RAPIDS also significantly outperformed single-modality prediction models (AUC 0·630 [0·507–0·754] for the pathomics microenvironment model, 0·716 [0·580–0·852] for the radiomics MRI model, and 0·733 [0·620–0·845] for the pathomics nucleus model; all p<0·0001).RAPIDS was able to predict pathological complete response to neoadjuvant chemoradiotherapy based on pretreatment radiopathomics images with high accuracy and robustness and could therefore provide a novel tool to assist in individualised management of locally advanced rectal cancer.National Natural Science Foundation of China; Youth Innovation Promotion Association of the Chinese Academy of Sciences.