Feldman, Michael D.

ImportancePathologic complete response (pCR) after neoadjuvant chemotherapy (NAC) in breast cancer strongly correlates with overall survival and has become the standard end point in neoadjuvant trials. However, there is controversy regarding whether the definition of pCR should exclude or permit the presence of residual ductal carcinoma in situ (DCIS).ObjectiveTo examine the association of residual DCIS in surgical specimens after neoadjuvant chemotherapy for breast cancer with survival end points to inform standards for the assessment of pathologic complete response.Design, Setting, and ParticipantsThe study team analyzed the association of residual DCIS after NAC with 3-year event-free survival (EFS), distant recurrence-free survival (DRFS), and local-regional recurrence (LRR) in the I-SPY2 trial, an adaptive neoadjuvant platform trial for patients with breast cancer at high risk of recurrence. This is a retrospective analysis of clinical specimens and data from the ongoing I-SPY2 adaptive platform trial of novel therapeutics on a background of standard of care for early breast cancer. I-SPY2 participants are adult women diagnosed with stage II/III breast cancer at high risk of recurrence.InterventionsParticipants were randomized to receive taxane and anthracycline-based neoadjuvant therapy with or without 1 of 10 investigational agents, followed by definitive surgery.Main Outcomes and MeasuresThe presence of DCIS and EFS, DRFS, and LRR.ResultsThe study team identified 933 I-SPY2 participants (aged 24 to 77 years) with complete pathology and follow-up data. Median follow-up time was 3.9 years; 337 participants (36%) had no residual invasive disease (residual cancer burden 0, or pCR). Of the 337 participants with pCR, 70 (21%) had residual DCIS, which varied significantly by tumor-receptor subtype; residual DCIS was present in 8.5% of triple negative tumors, 15.6% of hormone-receptor positive tumors, and 36.6% of ERBB2-positive tumors. Among those participants with pCR, there was no significant difference in EFS, DRFS, or LRR based on presence or absence of residual DCIS.Conclusions and RelevanceThe analysis supports the definition of pCR as the absence of invasive disease after NAC regardless of the presence or absence of DCIS.Trial RegistrationClinicalTrials.gov Identifier NCT01042379.

Abstract Patients with Hodgkin lymphoma (HL) who relapse or are refractory to first-line therapy have a poor prognosis and some of these patients require innovative therapeutic approaches. Our group and others have demonstrated the clinical efficacy of anti-CD19 chimeric antigen receptor T cells (CART19) for B cell malignancies. However, B cell antigens such as CD19 are typically absent from Hodgkin Reed-Sternberg (HRS) cells. Targeting the CD30 antigen using CART cells has met with limited clinical success to date. In HL the neoplastic HRS cells often represent ~1% of the tumor mass and the remainder is comprised of a highly immunosuppressive tumor microenvironment (TME). Within the TME, tumor-associated macrophages (TAM) have a key role in promoting tumor growth and inhibiting the anti-tumor immune response, and the presence of TAM in HL is associated with a poor prognosis. HL therefore represents an excellent model of tumor-induced immunosuppression that can potentially represent a barrier to successful CART therapy. The objective of this study was therefore to develop an anti-HL CAR T cell immunotherapy that would also overcome the immunosuppression of the HL microenvironment. We sought to define a tumor antigen that was co-expressed on HRS and on TAM. Using immunohistochemistry and flow cytometry we confirmed previous reports that CD123 (the IL-3Rα chain) is expressed on HRS in 50-60% of HL patients and in 100% HL cell lines (HDLM-2, KMH2, SUPHD1, and L428). Importantly, CD123 was also seen on several HL-associated immune cells of the TME, including TAM. Exposure of macrophages to HDLM-2 cells polarized them to an M2-like phenotype (Figure 1 A) with expression of CD123. Luminex analysis of HDLM-2/TAM co-culture supernatants revealed high levels of IL-13, and blockade of IL-13 partially reversed the immunosuppressive phenotype. We then developed a model of TAM-mediated CAR T cell inhibition, using immunosuppressive M2-like macrophages derived by exposure to IL-4 or to HDLM-2 cells. These macrophages could suppress “third party” CART19 proliferation in response to B-acute lymphoblastic leukemia. We then used anti-CD123 CART cells (CART123) to target TAM. Importantly, CART123 actively recognized and killed TAM, thereby resisting TAM-mediated inhibition (as opposed to control CART19 cells) (Figure 1 B). We therefore sought to investigate the utility of anti-CD123 CART for the treatment of HL (initial results for this part presented at ASH 2014). In vitro, CART123 specifically degranulate, proliferate, produce cytokines and kill HL cells. A clinically-relevant xenograft model of HL was generated using luciferase+ HDLM-2 cells in NSG mice. In this model mice showed progressive increase in tumor burden over approximately 6 weeks, reproducing the indolent nature of the human disease. At day ~43 mice were randomized to receive 1.5 million CART123 cells or control T cells. CART123 induced complete and durable eradication of disseminated tumor within 14 days, leading to 100% relapse-free and 100% overall survival at 8 months while mice treated with control T cells had a median survival of 128 days (Figure 2 A). Tumor elimination was associated with extensive CAR T cell expansion as detected by flow cytometry in serial peripheral blood bleedings. In order to prove the formation of an immunological memory, at day 250 CART123 treated mice were re-challenged with the HDLM-2 cell line. The tumor was promptly rejected, associated with a re-expansion of CART123 cells in the peripheral blood (~10 months after T cell injection) leading to an overall survival advantage as compared to controls (Figure 2 B). In summary, we showed that human CD123-redirected T cells display potent therapeutic activity against disseminated HL. In Hodgkin Lymphoma (HL) CD123 is expressed on malignant Reed-Sternberg cells as well as in M2-polarized tumor associated macrophages (TAMs): importantly CART123 can target simultaneously both the HRS and the TAM without being resistant to immunosuppression. We have previously demonstrated that CART123 can lead to myelosuppression in preclinical models, suggesting that our findings could be translated to treat patients with refractory HL in the context of stem cell transplantation using “short-acting” RNA-CAR123 T cells or depletable lentivirally-transduced CART123. Disclosures Ruella: Novartis: Patents & Royalties. Kenderian:Novartis: Patents & Royalties, Research Funding. Wasik:Gilead Sciences: Equity Ownership; Seattle Genetics: Honoraria; Novartis: Research Funding; University of Pennsylvania: Patents & Royalties: NPM-ALK as an omncogene; University of Pennsylvania: Patents & Royalties: CAR T-cells; Gilead Sciences: Research Funding; Pharmacyclics: Research Funding. June:Novartis: Honoraria, Patents & Royalties, Research Funding; Johnson & Johnson: Honoraria; Pfizer: Honoraria; Celldex: Consultancy, Equity Ownership; Immune Design: Consultancy, Equity Ownership; Tmunity Therapeutics: Equity Ownership; Novartis: Honoraria, Patents & Royalties, Research Funding. Gill:Novartis: Patents & Royalties, Research Funding.

Abstract Background: Pathological complete response (pCR) is accepted by FDA as a surrogate endpoint for accelerated approval of targeted agents in combination with chemotherapy based on better long-term outcomes compared to residual disease (Cortazar 2014). Methods: The multi-center, adaptively-randomized I-SPY2 platform trial uses pCR as the primary endpoint to identify investigational agents that will improve outcomes in women with stage 2/3 breast cancer with high risk of early recurrence, across all signatures, based on hormone receptor (HR), HER2, and 70-gene (MammaPrint) status. For patients with HR+ HER2- tumors, only 70-gene (Mammaprint) high-risk patients are enrolled. To date, 1200+ patients have been randomized to one of 14 arms: control (paclitaxel followed by AC); veliparib/carboplatin; neratinib; MK2206; trebananib; trastuzumab/pertuzumab; ado-trastuzumab emtansine/pertuzumab; pembrolizumabx4; ganitumab/metformin; ganetespib; PLX-3397. 7 agents graduated in at least one signature (> 85% probability of success in a 300-patient phase III confirmatory trial); 2 did not graduate; 1 stopped for toxicity, and 3 are enrolling (patritumab/trastuzumab, talazoparib/irinotecan, pembrolizumabx8). Local pathologists were centrally trained using the Residual Cancer Burden (RCB) assessment to ensure uniform evaluation and response classification; RCB 0 = pCR. Results: We evaluated the relationship between pCR and event free (EFS) and distant disease free survival (DDFS) in the first 522 pts (median follow-up:2.5 years). 180 pts achieved pCR (36%) while 338 did not (RCB=1-3). There were 82 EFS and 65 DRFS events. Over the entire group (including all arms), pCR was highly associated with 3-year EFS (p<0.001 for both). Pts achieving pCR had a 3% recurrence risk (RR) at 3 years; those with non-pCR had 24% RR over this time period. For distant recurrence, the 3-year RR with pCR was 2%, compared to 20% in pts with non-pCR. As expected, pCR rates varied by breast cancer subtype (HR+/HER2: 18% (35/196), HR+/HER2+: 40% (33/82), HR-/HER2+:68% (34/50), HR-/HER2-:41% (76/188)). The relationship between pCR and EFS was significant and clinically impactful within each subtype. 3-year survival (pCR group)Hazard Ratio OverallOverallHR+/HER2-HER2+TNBCEFS97%0.08 (0.03-0.23)0.14 (0.02-1.04)0 (NA)0.11 (0.03-0.37)DDFS98%0.08 (0.03-0.26)0.17 (0.01-1.23)0 (NA)0.09 (0.02-0.40) Conclusions: The first long-term efficacy results from the I-SPY2 TRIAL demonstrate that achieving pCR is a very strong surrogate endpoint for improved EFS and DDFS in a high-risk population, across all treatment arms, regardless of subtype. I-SPY2 shows substantially lower estimated EFS hazards for patients achieving pCR, compared to the 5 yr EFS hazard ratio for pCR vs not in Cortazar (hazard ratio 0.49), demonstrating important differences between a metaanalysis compared to a platform trial with uniform high-risk eligibility, standardized pathology assessment, and multiple targeted therapies. Our data support the use of pCR as a primary endpoint for accelerated approval of new drugs if EFS is evaluated in the same population. Based on these findings, the I-SPY2 TRIAL will test whether therapy can be deescalated or escalated for individual patients with the goal of achieving pCR for all. Citation Format: Yee D, DeMichele A, Isaacs C, Symmans F, Yau C, Albain KS, Hylton NM, Forero-Torres A, van't Veer LJ, Perlmutter J, Rugo HS, Melisko M, Chen Y-Y, Balassanian R, Krings G, Datnow B, Hasteh F, Tipps A, Weidner N, Zhang H, Tickman R, Thornton S, Ritter J, Amin K, Klein M, Chen B, Keeney G, Ocal T, Feldman M, Klipfel N, Sattar H, Mueller J, Gwin K, Baker G, Kallakury B, Zeck J, Duan X, Ersahin C, Gamez R, Troxell M, Mansoor A, Grasso LeBeau L, Sams S, Wisell J, Wei S, Harada S, Vinh T, Stamatakos MD, Tawfik O, Fan F, Adams A, Rendi M, Minton S, Magliocco A, Sahoo S, Fang Y, Hirst G, Singhrao R, Asare SM, Wallace AM, Chien AJ, Ellis ED, Han HS, Clark AS, Boughey JC, Elias AD, Nanda R, Korde L, Murthy R, Lang J, Northfelt D, Khan Q, Edmiston KK, Viscusi R, Haley B, Kemmer K, Zelnak A, Berry DA, Esserman LJ. Pathological complete response predicts event-free and distant disease-free survival in the I-SPY2 TRIAL [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr GS3-08.

Abstract Patients with otherwise treatment-resistant Hodgkin lymphoma could benefit from chimeric antigen receptor T-cell (CART) therapy. However, Hodgkin lymphoma lacks CD19 and contains a highly immunosuppressive tumor microenvironment (TME). We hypothesized that in Hodgkin lymphoma, CART should target both malignant cells and the TME. We demonstrated CD123 on both Hodgkin lymphoma cells and TME, including tumor-associated macrophages (TAM). In vitro, Hodgkin lymphoma cells convert macrophages toward immunosuppressive TAMs that inhibit T-cell proliferation. In contrast, anti-CD123 CART recognized and killed TAMs, thus overcoming immunosuppression. Finally, we showed in immunodeficient mouse models that CART123 eradicated Hodgkin lymphoma and established long-term immune memory. A novel platform that targets malignant cells and the microenvironment may be needed to successfully treat malignancies with an immunosuppressive milieu. Significance: Anti-CD123 chimeric antigen receptor T cells target both the malignant cells and TAMs in Hodgkin lymphoma, thereby eliminating an important immunosuppressive component of the tumor microenvironment. Cancer Discov; 7(10); 1154–67. ©2017 AACR. This article is highlighted in the In This Issue feature, p. 1047

Abstract Background: I-SPY2 is a multicenter phase 2 trial in high risk stage II/III breast cancer (BC) using adaptive randomization within biomarker subtypes to evaluate novel agents added to standard neoadjuvant chemotherapy. The first regimen to graduate based on the predicted probability of a higher pCR rate within predefined subsets was veliparib/carboplatin + paclitaxel (VC+T→AC vs T→AC) in triple negative BC (TNBC). In TNBC the residual cancer burden (RCB) is prognostic, whether as a continuous index or grouped into classes, with pCR (RCB-0) and RCB-I classes having identical survival. Therefore, we evaluated the use of RCB to further discriminate between investigational and control arms. Methods: Site pathologists reported RCB for 99% of subjects in the primary efficacy analysis based on pCR (n=114/115). We compared the distribution of RCB reported as a continuous index in each treatment-subset combination to matched concurrently randomized controls using the Wilcoxon rank sum test for RCB index, and Fisher's Exact test for RCB classes (RCB-0/I vs RCB-II/III). The statistics are descriptive rather than inferential, and given the small sample size have no claim on generalizability. We modified the Bayesian model used to compute the estimated probability of success in a future, randomized, phase 3 trial of 300 subjects, if response were defined by either pCR or RCB-I (RCB0/I), or separately if it were defined by pCR alone. Results: VC+T→AC led to a significantly lower RCB index than T→AC in TNBC (p=0.0021), with a near-significant trend when those with pCR were excluded (p=0.06). There was no significant difference in RCB distributions in the other breast cancer subtypes treated. In TNBC, the odds ratio (OR) for achieving RCB-0/I in the VC+T→AC arm vs control was 8.2 (95% confidence interval (CI): 2.1–35), whereas the OR for achieving pCR was 4.56 (95% CI: 1.25–19.53). The simulations using response information from I-SPY2 to predict the probability of success for VC+T→AC for TNBC in a future phase 3 trial estimated this probability to be 0.99 if modeled using RCB-0/I as the response endpoint, and 0.90 if modeled using pCR as the response endpoint. Conclusions: Use of RCB index and classes provided additional insight into the effect of adding VC to T, appearing to magnify the improved treatment response that had been observed with pCR rates in TNBC. It will be important to test in randomized trials whether a decrease in the RCB index relative to controls, and/or increased rates of RCB-0/I class, are predictive of survival benefit in TNBC. Citation Format: Liu MC, Symmans WF, Yau C, Chen Y-Y, Rugo HS, Olopade OF, Datnow B, Chen B, Feldman M, Kallakury B, Hasteh F, Tickman R, Ritter J, Troxel M, Mhawech-Fauceglia P, Duan X, Berry D, Esserman L, DeMichele A. Residual cancer burden (RCB) with veliparib/carboplatin in the I-SPY2 trial. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-07-49.

Introduction: The advent of digital slides offers new opportunities within the practice of pathology such as the use of image analysis techniques to facilitate computer aided diagnosis (CAD) solutions. Use of CAD holds promise to enable new levels of decision support and allow for additional layers of quality assurance and consistency in rendered diagnoses. However, the development and testing of prostate cancer CAD solutions requires a ground truth map of the cancer to enable the generation of receiver operator characteristic (ROC) curves. This requires a pathologist to annotate, or paint, each of the malignant glands in prostate cancer with an image editor software - a time consuming and exhaustive process.Recently, two CAD algorithms have been described: probabilistic pairwise Markov models (PPMM) and spatially-invariant vector quantization (SIVQ). Briefly, SIVQ operates as a highly sensitive and specific pattern matching algorithm, making it optimal for the identification of any epithelial morphology, whereas PPMM operates as a highly sensitive detector of malignant perturbations in glandular lumenal architecture.Methods: By recapitulating algorithmically how a pathologist reviews prostate tissue sections, we created an algorithmic cascade of PPMM and SIVQ algorithms as previously described by Doyle el al. [1] where PPMM identifies the glands with abnormal lumenal architecture, and this area is then screened by SIVQ to identify the epithelium.Results: The performance of this algorithm cascade was assessed qualitatively (with the use of heatmaps) and quantitatively (with the use of ROC curves) and demonstrates greater performance in the identification of malignant prostatic epithelium.Conclusion: This ability to semi-autonomously paint nearly all the malignant epithelium of prostate cancer has immediate applications to future prostate cancer CAD development as a validated ground truth generator. In addition, such an approach has potential applications as a pre-screening/quality assurance tool.