E. Dmitrovsky

The Influence of Body Mass Index on Overall Survival Following Surgical Resection of Non–Small Cell Lung Cancer

Background Population studies suggest that high body mass index (BMI) correlates with a reduced risk for death from lung cancer. The aim of our study was to evaluate definitively the influence of BMI on long‐term overall survival (OS) in surgical patients with NSCLC. Methods The study population consisted of 1935 patients who underwent surgical resection for lung cancer at M. D. Anderson Cancer Center (2000–2014). Study variables included both patient‐ and treatment‐related characteristics. Univariate and multivariate Cox regression analyses were performed to identify variables associated with OS. Results On univariate analysis, significant predictors of improved survival were higher BMI, pathologic tumor stage (stage I versus stage II, III, or IV), type of surgical procedure (lobectomy/pneumonectomy versus wedge resection/segmentectomy), younger age, female sex, and adenocarcinoma histologic subtype (versus squamous) (all p < 0.05). Morbidly obese patients (BMI ≥ 35) demonstrated a trend toward better outcomes than those classified as obese (BMI ≥30 and <35 kg/m2) (p = 0.05), overweight (BMI ≥ 25 and <30 kg/m2) (p = 0.13), or healthy weight (BMI <25 kg/m2) (p = 0.37) (hazard ratio = 0.727, 0.848, 0.926, and 1, respectively). On multivariate analysis, BMI remained an independent predictor of survival (p = 0.02). Propensity matching analysis demonstrated significantly better OS (p = 0.003) in patients with a BMI of 30 kg/m2 or higher as compared with a BMI of 25 kg/m2. Conclusions In this large, retrospective, single‐center series, after control for disease stage and other variables, higher BMI was associated with improved OS after surgical resection of NSCLC. Further studies are necessary to elucidate the precise relationship between BMI and treatment outcomes.

G0S2 suppresses oncogenic transformation by repressing a Myc-regulated transcriptional program

Methylation-mediated silencing of G0-G1 switch gene 2 (G0S2) has been detected in a variety of solid tumors, whereas G0S2 induction is associated with remissions in patients with acute promyelocytic leukemia, implying that G0S2 may possess tumor suppressor activity. In this study, we clearly demonstrate that G0S2 opposes oncogene-induced transformation using G0s2-null immortalized mouse embryonic fibroblasts (MEF). G0s2-null MEFs were readily transformed with HRAS or EGFR treatment compared with wild-type MEFs. Importantly, restoration of G0S2 reversed HRAS-driven transformation. G0S2 is known to regulate fat metabolism by attenuating adipose triglyceride lipase (ATGL), but repression of oncogene-induced transformation by G0S2 was independent of ATGL inhibition. Gene expression analysis revealed an upregulation of gene signatures associated with transformation, proliferation, and MYC targets in G0s2-null MEFs. RNAi-mediated ablation and pharmacologic inhibition of MYC abrogated oncogene-induced transformation of G0s2-null MEFs. Furthermore, we found that G0S2 was highly expressed in normal breast tissues compared with malignant tissue. Intriguingly, high levels of G0S2 were also associated with a decrease in breast cancer recurrence rates, especially in estrogen receptor-positive subtypes, and overexpression of G0S2 repressed the proliferation of breast cancer cells in vitro. Taken together, these findings indicate that G0S2 functions as a tumor suppressor in part by opposing MYC activity, prompting further investigation of the mechanisms by which G0S2 silencing mediates MYC-induced oncogenesis in other malignancies.

CD38-mediated immunosuppression as a mechanism of tumor cell escape from PD-1/PD-L1 blockade

Although treatment with immune checkpoint inhibitors provides promising benefit for patients with cancer, optimal use is encumbered by high resistance rates and requires a thorough understanding of resistance mechanisms. We observed that tumors treated with PD-1/PD-L1 blocking antibodies develop resistance through the upregulation of CD38, which is induced by all-trans retinoic acid and IFNβ in the tumor microenvironment. In vitro and in vivo studies demonstrate that CD38 inhibits CD8+ T-cell function via adenosine receptor signaling and that CD38 or adenosine receptor blockade are effective strategies to overcome the resistance. Large data sets of human tumors reveal expression of CD38 in a subset of tumors with high levels of basal or treatment-induced T-cell infiltration, where immune checkpoint therapies are thought to be most effective. These findings provide a novel mechanism of acquired resistance to immune checkpoint therapy and an opportunity to expand their efficacy in cancer treatment.Significance: CD38 is a major mechanism of acquired resistance to PD-1/PD-L1 blockade, causing CD8+ T-cell suppression. Coinhibition of CD38 and PD-L1 improves antitumor immune response. Biomarker assessment in patient cohorts suggests that a combination strategy is applicable to a large percentage of patients in whom PD-1/PD-L1 blockade is currently indicated. Cancer Discov; 8(9); 1156-75. ©2018 AACR.See related commentary by Mittal et al., p. 1066This article is highlighted in the In This Issue feature, p. 1047.

The Retinoids and Cancer Chemoprevention

The retinoids—natural and synthetic derivatives of vitamin A—play a role in cancer therapy and prevention. Their role in chemoprevention has been highlighted through the results of clinical and epidemiological studies in conjunction with basic research on retinoid mechanisms of action, advances in genomic information, novel organic synthesis of retinoid analogs, and availability of reliable model systems. This chapter focuses on the results of recent retinoid clinical cancer therapy and chemoprevention trials, reviews the current understanding of the molecular mechanism of retinoid action, discusses in vitro models used to study retinoid chemoprevention, and summarizes information on retinoid target genes. An understanding of the convergence of clinical and basic scientific advances in the retinoid field should lead to improved strategies to use these pharmacological agents in cancer therapy or chemoprevention.

Reply to “Influence of Body Mass Index on Overall Survival Following Surgical Resection of Non–Small Cell Lung Cancer: Methodological Issues”

Figure 1. Log minus log function for patterns 1 and 2. LML, log minus log; surgdate, date of surgical procedure. In Response: We appreciate the interest of Safiri et al. in our recent publication titled “Influence of Body Mass Index on Overall Survival Following Surgical Resection of Non–Small Cell Lung Cancer.” Our large retrospective single-center study revealed that improved overall survival was associated with higher body mass index (BMI) after surgical resection of NSCLC. Perioperative BMI was routinely measured before surgery. Our institutional database showed that patients with lung cancer having surgical resection did not typically lose weight between the time of their initial visit and the operation. Although within-person variability of BMI exists in other medical conditions, it was not observed in this case. For this reason, we view our measurement of BMI as reliable. As to the possible concern regarding a residual confounder, the multivariate analysis examined all variables that were included in univariate analysis, and the conclusions drawn were similar in the different analyses. A retrospective study might not eliminate the possibility of a residual confounder. Yet, only clinically available variables were considered in this retrospective study. As such, only a future controlled clinical trial would be able eliminate the possibility of a residual confounder. We did conduct a log minus log plot of survival to test the validity of the regression model that was used. The curves provided parallel lines, as displayed in Figure 1. This revealed that the assumption of the proportional hazards model is valid and in compliance with a rigorous statistical assessment. Given this, we view our provided analyses as sound and thank Safiri et al. for their interest in our work. Boris Sepesi, MD, Arlene M. Correa, PhD Department of Thoracic and Cardiovascular Surgery University of Texas M. D. Anderson Cancer Center Houston, Texas

Abstract 24: Loss of ubiquitin-specific peptidase 18 (USP18) causes cold sensitive mice by destabilizing the critical regulator of thermogenesis: uncoupling protein-1 (UCP-1)

USP18 is the major ISG15 (Interferon-Stimulated Gene 15) deconjugase that removes ISG15 from substrate proteins. We recently reported that USP18 null mice spontaneously develop leiomyosarcomas. Intriguingly, these USP18 null mice are also markedly cold sensitive as compared to their wild-type littermates. When briefly exposed to cold stimuli, USP18 null mice significantly (P Citation Format: Xi Liu, Yun Lu, Lin Zheng, David J. Sekula, Sarah J. Freemantle, Ethan Dmitrovsky. Loss of ubiquitin-specific peptidase 18 (USP18) causes cold sensitive mice by destabilizing the critical regulator of thermogenesis: uncoupling protein-1 (UCP-1). [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 24.

Abstract 1255: Loss of the ubiquitin protease USP18 represses KRAS mutant lung cancer tumorigenicity in mice by destabilizing KRAS protein

KRAS is frequently mutated in lung cancers. Innovative strategies are needed to combat KRAS mutant lung cancers because these tumors are often resistant to therapy. This study reports that loss of the deubiquitinase USP18 leads to destabilization of the KRAS oncogenic protein in panels of murine and human lung cancer cells. In marked contrast, engineered gain of USP18 expression using retroviral vectors introduced into the same panel of murine and human lung cancer cells stabilized KRAS protein. Loss of USP18 expression in KRAS mutant-expressing lung cancer cells inhibited their growth while gain of USP18 expression opposed this effect. We sought to identify mechanisms engaged in this KRAS protein destabilization. Intriguingly, immunoprecipitation assays established that KRAS conjugates with the ubiquitin-like protein ISG15. This leads to KRAS protein destabilization. Using the protein synthesis inhibitor cycloheximide, USP18 knockdown was shown to destabilize mutant KRAS more prominently than wild-type KRAS protein. To independently confirm that KRAS directly complexes with ISG15, site-directed mutagenesis was performed to render the C-terminal domain of KRAS lysine-less. This led to stabilization of KRAS, despite knockdown of USP18. These studies were confirmed and extended in the in vivo setting of KRAS-driven lung cancers in KRASLA2/+ mice within the FVB strain background. We crossed these mice with USP18 null (USP18-/-) mice to obtain KRASLA2/+;USP18-/- compound mice. Strikingly, these compound mice had statistically significantly (P Citation Format: Lisa Maria Mustachio, Yun Lu, Laura J. Tafe, Angeline S. Andrew, Vincent Memoli, Jaime Rodriguez-Canales, Pamela A. Villalobos, Ignacio Wistuba, Jun Yu, Jack J. Lee, Fadzai Chinyengetere, David J. Sekula, Xi Liu, Sarah J. Freemantle, Ethan Dmitrovsky. Loss of the ubiquitin protease USP18 represses KRAS mutant lung cancer tumorigenicity in mice by destabilizing KRAS protein. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1255.

Abstract CN05-03: Cooperation between a rexinoid and EGFR-TKI for lung cancer prevention via Cyclin D1 destabilization

We previously reported that both nuclear retinoic acid receptor (RAR, retinoid) and retinoid X receptor (RXR, rexinoid) agonists can trigger proteasomal degradation of cyclins. This confers check point arrest and repair of carcinogenic DNA damage in bronchial epithelial cells. Mechanisms responsible for this induced degradation were discovered. These included ubiquitin-dependent as well as ISG15-dependent programs that independently destabilized expression of cyclin D1 and other G1 cyclin proteins. The critical receptor that confers this cyclin destabilization was RARβ. Yet, silencing of RARβ; and specifically of the previously unrecognized isoform that we cloned and designated as RARβ1 likely accounts for clinical resistance to classical retinoids (like 13-cis-retinoic acid and all-trans-retinoic acid) in lung carcinogenesis. We sought to learn whether RXR/RAR heterodimer complex activation with a rexinoid was able to trigger cyclin destabilization. This was found to be the case in in vitro studies. This finding implied that the same pathway would be engaged in the in vivo setting. To establish if this occurs, we engineered transgenic mice independently with human surfactant C-driven wild-type cyclin E or a proteasome-degradation resistant cyclin E species. This mouse model was developed because human pre-malignant and malignant lung lesions frequently deregulate cyclin expression. Intriguingly, these mice recapitulated many features of lung carcinogenesis found in patients. Neoplastic changes were enhanced by transgenic expression of the degradation-resistant cyclin E species. We built on this finding by showing that a rexinoid and an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) each chemoprevented lung cancers in vinyl carbamate-induced lung tumors within strain A/J mice. Notably, the rexinoid was more potent that the EGFR-TKI in conferring this chemoprevention. Both agents also reduced cyclin expression, but did so through distinct mechanisms. We sought to learn if a rexinoid (bexarotene) cooperated with an EGFR-TKI (erlotinib) in exerting anti-tumorigenic effects in lung cancer. Cooperation between these agents was found in both murine and human lung cancer cells and even in those cells that exhibited KRAS/p53 mutations. Over the past decade, we translated this work into the cancer clinic through a series of five clinical trials that moved this work through successive phase 0, phase I, and phase II trials of lung and aerodigestive tract cancers. In window of opportunity trials, pharmacodynamic responses (cyclin repression and induced necrosis and inflammatory responses) were seen when intratumoral levels of these drugs were comparable to those necessary to trigger in vitro effects. A phase II trial in heavily pre-treated stage IV non-small cell lung cancer (NSCLC) cases was performed. EGFR and cyclin expression profiles as well as KRAS mutations were searched for in these NSCLC cases. Findings revealed substantial cyclin repression and reduction of lung cancer growth by combined therapy with a rexinoid (bexarotene) and EGFR-TKI (erlotinib). Cyclins were repressed while necrosis and inflammation were induced in post-treatment versus pre-treatment lung tumor biopsies obtained in this window of opportunity trial. Objective anti-tumor responses occurred whether or not KRAS or activating EGFR mutations were detected. This refractory NSCLC trial had 3 major clinical responses (2 had KRAS or EGFR mutations) with prolonged survival (583, 665, and 1460+ days, respectively). Median survival was 22 weeks (16 weeks for controls). Hypertriglyceridemia or rash significantly increased median overall survival to 24 weeks. Thus, this combination regimen revealed substantial clinical anti-tumor activity against NSCLCs. Taken together, these findings indicate that cooperation between a rexinoid and EGFR-TKI can chemoprevent and treat lung cancers by causing cyclin destabilization. These agents are also useful tools to identify other antineoplastics that repress lung cancers by destabilizing cyclin expression or by inhibiting their associated kinases. Evidence for this will be presented in this session. Citation Format: Konstantin H. Dragnev, Vincent Memoli, Sarah J. Freemantle, Samuel Waxman, Ethan Dmitrovsky. Cooperation between a rexinoid and EGFR-TKI for lung cancer prevention via Cyclin D1 destabilization. [abstract]. In: Proceedings of the Thirteenth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2014 Sep 27-Oct 1; New Orleans, LA. Philadelphia (PA): AACR; Can Prev Res 2015;8(10 Suppl): Abstract nr CN05-03.

Abstract ED05-03: Rexinoid nuclear receptors and cancer prevention

This session will present on the use of retinoid X receptor (RXR) agonists (rexinoids) for cancer prevention. Several areas will be highlighted. Using lung cancer as a model, both preclinical and clinical evidence for use of RXR-selective ligands in cancer prevention will be discussed. Emphasis will be placed on the pharmacology of newer rexinoids. Lung cancer is the most common cause of cancer death for men and women. Innovative ways to combat lung cancer are needed. A previously unrecognized antineoplastic pathway that triggers proteasomal degradation of different cyclins was found. Both retinoic acid receptor (RAR) and RXR agonists (among other agents) cause degradation of cyclins. This confers check point arrest and repair of carcinogenic damage of DNA in bronchial epithelial cells. Mechanisms responsible for this induced proteasomal degradation will be presented. The critical receptor that confers cyclin destabilization is RARβ. Yet, silencing of RARβ and of the novel isoform RARβ19 likely accounts for clinical resistance to classical retinoids in lung carcinogenesis. It was explored whether RXR/RAR heterodimer complex activation with a rexinoid triggers G1 cyclin destabilization. This was found and this result implied that the same pathway would be engaged in vivo. To establish if this occurred in the lung, transgenic mice were independently engineered with human surfactant C-driven wild-type cyclin E or a proteasome-degradation resistant cyclin E species. This was done because human pre-malignant and malignant lung lesions deregulate cyclin expression. Intriguingly, these mice recapitulated frequent features of lung carcinogenesis found in patients. These changes were enhanced by the expression of degradation-resistant cyclin E species. This finding was built upon by showing that a rexinoid can chemoprevent lung cancers in vinyl carbamate-induced lung tumors within strain A/J mice. The epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI) erlotinib also repressed cyclin D1 expression, but through different mechanisms. The rexinoid bexarotene was combined with erlotinib. Effects on transgenic lung cancer cell growth and cyclin D1 expression were studied when KRAS or KRAS/p53 mutations were present or not. Over the past decade, this work was confirmed by extending analyses into the cancer clinic through a series of five clinical trials that moved this work through successive phase 0, phase I, and phase II trials. In window of opportunity trials, pharmacodynamic responses (cyclin repression and induced necrosis and inflammatory responses) were seen when intratumoral drug levels were comparable to those detected in vitro. Cyclin D1 levels were measured in pre-treatment versus post-treatment tumor biopsies. A phase II trial in heavily pre-treated stage IV non-small cell lung cancer (NSCLC) cases was performed. EGFR expression and KRAS mutations were examined in these cases. Findings revealed substantial cyclin repression and reduction of lung cancer growth by combined treatment with a rexinoid and EGFR-TKI. Cyclins were repressed while necrosis and inflammatory responses were induced in the window of opportunity trial independent of KRAS or activating EGFR mutations. This refractory NSCLC trial had 3 major clinical responses (2 had KRAS or EGFR mutations) with prolonged survival, as will be discussed. Median survival was 22 weeks (16 weeks for controls). Hypertriglyceridemia or rash significantly increased median overall survival to 24 weeks. This regimen revealed anti-tumor activity independent of the presence of EGFR or KRAS mutations in NSCLCs. Notably, these findings were independently replicated in the lung cancer BATTLE trial. An analogous approach was used to uncover a distinct rexinoid-induced protein destabilization pathway that involves the deubiquitinase USP18/UBP43, as will be discussed. Taken together, these findings reveal that rexinoid-induced cyclin destabilization is a previously unrecognized antineoplastic pathway. Thus, additional rexinoid trials for cancer therapy and prevention are warranted. Citation Format: Ethan Dmitrovsky. Rexinoid nuclear receptors and cancer prevention. [abstract]. In: Proceedings of the Twelfth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2013 Oct 27-30; National Harbor, MD. Philadelphia (PA): AACR; Can Prev Res 2013;6(11 Suppl): Abstract nr ED05-03.

A window-of-opportunity trial of vorinostat (V) in patients with resectable non-small cell lung cancer (NSCLC).

7062 Background: Lung cancer is the most common cause of cancer mortality in the United States and novel treatments are needed. We previously engineered murine transgenic cyclin E models that recapitulated frequent features of human lung cancer including chromosome instability, hedgehog pathway, cyclin D1 activation and pre-malignant and malignant lung lesions. Cell lines (ED-1 and ED-2) were derived from these lung cancers as models to identify promising anti-neoplastic agents. Among a panel of drugs that caused growth inhibition of lung cancer cells and repression of cyclin D1, the histone deacetylase inhibitor vorinostat (V) was found to be most potent. V repressed cyclin D1 and cyclin E proteins, but increased p27 expression in ED-1 cells. V treatment also significantly reduced lung tumor formation in mice harboring transplanted syngeneic lung cancer cells. To translate studies into the clinic, a window of opportunity trial of V was conducted in patients (pts) with early stage NSCLC.nnnMETHODSnPts with resectable NSCLC received 400 mg V once daily orally for 7 days prior to undergoing surgical resection. Post- versus pre-treatment tumor biopsies were scored for changes in immunohistochemical profiles of cyclin D1, cyclin E, EGFR, phospho-EGFR and Ki-67.nnnRESULTSnTen pts were enrolled, median age 66 (60-73) years, 60% were women, 1 was a never smoker, 4 were current smokers, pathological stages IA-IIB, 2 had squamous cell carcinoma, 7 had adenocarcinoma, 1 had poorly differentiated carcinoma. Five pts completed the study treatment, 3 pts received V for less than 7 days, 2 did not take the study medication. Ki-67 staining decreased in all examined pairs, cyclin D1 levels decreased in 1 sample, 3 pts had reduced cyclin E expression, 2 pts had reduced EGFR expression, and there was no change in phospho-EGFR detection. Four pairs showed necrosis in the posttreatment specimens. Results for p27, other exploratory markers, and dynamic CT will be reported at the meeting.nnnCONCLUSIONSnPreoperative treatment with V decreased Ki-67 and other biomarkers in resected NSCLC. The results from this window of opportunity trial demonstrate the value of clinically-relevant mouse lung cancer models to guide NSCLC clinical trials.