Impact of Radiation Dose on Postoperative Complications in Esophageal and Gastroesophageal Junction Cancers

Introduction: The impact of radiation prescription dose on postoperative complications during standard of care trimodality therapy for operable stage II-III esophageal and gastroesophageal junction cancers has not been established. Methods: We retrospectively reviewed 82 patients with esophageal or gastroesophageal junction cancers treated between 2004 and 2016 with neoadjuvant chemoradiation followed by resection at a single institution. Post-operative complications within 30 days were reviewed and scored using the Comprehensive Complication Index (CCI). Results were compared between patients treated with <50 Gy and ≥ 50 Gy, as well as to published CROSS study neoadjuvant chemoradiation group data (41.4 Gy). Results: Twenty-nine patients were treated with <50 Gy (range 39.6-46.8 Gy) and 53 patients were treated with ≥ 50 Gy (range 50.0-52.5 Gy) delivered using IMRT/VMAT (41%), 3D-CRT (46%), or tomotherapy IMRT (12%). Complication rates and CCI scores between our <50 Gy and ≥ 50 Gy groups were not significantly different. Assuming a normal distribution of the CROSS data, there was no significant difference in CCI scores between the CROSS study neoadjuvant chemoradiation, <50 Gy, or ≥ 50 Gy groups. Rates of pulmonary complications were greater in the CROSS group (50%) than our <50 Gy (38%) or ≥ 50 Gy (30%) groups. Conclusions: In selected esophageal and gastroesophageal junction cancer patients, radiation doses ≥ 50 Gy do not appear to increase 30 day post-operative complication rates. These findings suggest that the use of definitive doses of radiotherapy (50-50.4 Gy) in the neoadjuvant setting may not increase post-operative complications.

Abstract P5-14-20: Mechanisms of radiation-induced brain edema alone and in combination with T-DM1 in Her2+ brain metastases

Introduction: Brain metastases (BM) occurs in ~40% of breast cancer patients with Her2+ tumors, which are typically treated with the Her-2 targeting drugs ado-trastuzumab emtansine (T-DM1) and trastuzumab. Radiotherapy (RTx), including stereotactic radiosurgery (SRS), is often utilized to treat BM. A subset of BM patients treated with SRS develops brain edema and radionecrosis (ERN), causing neurological deterioration, functional loss, and rarely death. We have reported an increased rate of ERN in patients treated with T-DM1 and SRS (39.1%), compared to those receiving SRS alone (4.5%). Thus, defining the mechanisms underlying ERN alone or in combination with T-DM1, is critical to reduce life-threatening side effects. While brain edema associated with BM and RTx is often attributed to blood-brain-barrier (BBB) disruption (vasogenic edema), recent studies now identify cytotoxic edema (osmotic cell expansion) as the initial andreversible step in a sequence of changes that leads to ionic edema, vasogenic edema, and ultimately brain swelling and necrosis. Here, we test the hypothesis that RTx alone and in combination with T-DM1 induces cytotoxic edema in astrocytes via upregulation of the water channel Aquaporin 4 (AQP4) and that blockage of cytotoxic edema will decrease ERN. Results: A cohort of BM from Her2+ patients treated with T-DM1/SRS showed enlarged reactive astrocytes expressing high levels of AQP4. To assess the role of astrocytic swelling in ERN, human astrocytes (THV) were treated with increasing doses of RTx and astrocytic cell size was measured 24 h later. A single 8 Gy RTx-dose increased cell size by 4.5 times compared to non-treated astrocytes (1248 ± 760.8 µm2 vs 5685 ± 4475 µm2, P&lt;0.0001). Consistent with a role for AQP4 in mediating astrocytic swelling, western blot (WB) showed increased AQP4 levels in RTx treated cells. To assess whether Her2-targeting (or directed) therapies impacted AQP4 expression and astrocytic survival/swelling, astrocytes were treated with increasing doses of RTx alone (0, 2, 4, 8 Gy), with trastuzumab or T-DM1 (1 μg/mL). WBs showed that only T-DM1 further exacerbated RTx-induced AQP4 upregulation and increased PARP cleavage. The combination of 4 Gy-RTx and T-DM1 showed a synergistic effect in reducing astrocytic confluence (32.1%), compared with astrocytes treated with 4 Gy-RTx alone (32.8 ± 3 vs 64.9 ± 13.3% respectively, P=0.0001, at 5 days). Moreover, T-DM1/RTx increased astrocytic size (P = 0.004) to a greater extent than equivalent RTx dose in combination with trastuzumab (P = 0.005). Other chemotherapeutic agents (paclitaxel and cisplatin) did not affect AQP4 protein levels, astrocytic survival or astrocytic swelling. TGN-020, a novel AQP4 inhibitor, and topiramate (TPM), an FDA-approved anticonvulsant/antiepileptic able to inhibit AQP4 activity, were evaluated for their potential role in reducing astrocytic swelling. Pretreatment with 100 µM TPM reduced RTx-induced astrocytic swelling by 36 % (7153 ± 4522 µm2 vs 4627 ± 3999 µm2, P&lt;0.0001). TPM pretreatment did not prevent 8 Gy-RTx-induced cell death (24.96 ± 5.7 % vs 23.47 ± 4.7 %) nor prevent DNA double stranded breaks measured by IF of p-γH2AX (89.17 ± 13.89 % vs 97.32 ± 2.94 %). Ongoing studies are assessing the effect of TPM on blocking astrocytic swelling and decreasing brain edema in vivo. Conclusions: These results suggest that astrocytic swelling (cytotoxic edema) is involved in the pathophysiology of ERN during BM treatment. In vitro, AQP4-inhibitors decreased astrocytic swelling without affecting astrocyte survival, suggesting that blockage of AQP4 could reduce brain edema without rescuing lethally-damaged astrocytes. Thus, a combination of drugs preventing vasogenic dysfunction (i.e bevacizumab) and AQP4-inhibitors (i.e TPM) could diminish ERN by targeting both, vasogenic and cytotoxic edema.

Citation Format: María J Contreras-Zárate, Steven Lai, Priscilla Stumpf, Christine Fisher, Ryan Ormond, Sana D Karam, Peter Kabos, Diana M Cittelly. Mechanisms of radiation-induced brain edema alone and in combination with T-DM1 in Her2+ brain metastases [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-14-20.

Impact of radiation dose during neoadjuvant chemoradiation on postoperative complications in esophageal (EC) and gastroesophageal junction cancers (GEJC)

119

Background: Neoadjuvant chemoradiation (nCRT) followed by resection is standard of care for operable stage II-III EC and GEJC; however, it can be associated with significant risk of postoperative complications (POC). The CROSS study group reported no increase in POC severity with nCRT using 41.4 Gy compared to surgery alone as defined by the Comprehensive Complication Index (CCI). We applied the CCI metric to evaluate the impact of nCRT radiation dose of < 50 Gy vs. ≥ 50 Gy on POC rates and compared to the CROSS rates. Methods: We retrospectively reviewed 82 pts (2004-2016) with EC or GEJC treated with nCRT followed by resection at our institution. 29 (35%) pts were treated with < 50 Gy (range 39.6-46.8 Gy) and 53 (65%) were treated with ≥ 50 Gy (range 50.0-52.5 Gy) delivered using IMRT/VMAT (41%), 3D-CRT (46%), or tomotherapy IMRT (12%). Concurrent chemotherapy were carboplatin/paclitaxel (59%), cisplatin/5-FU (17%), or other (24%). Resection was performed by Ivor Lewis esophagectomy (67%), esophagogastrectomy (14%), or trans-hiatal approach (11%). POC within 30 days were graded using the Clavien-Dindo scale and CCI scores were computed and compared between the two dose groups and with the CROSS nCRT group. Results: CCI scores and complication rates between our < 50 Gy and ≥ 50 Gy groups were not significantly different. Assuming a normal distribution of the CROSS CCI scores, there was no significant difference in CCI scores between the CROSS study nCRT, < 50 Gy, or ≥ 50 Gy groups. Rates of pulmonary complications were greater in the CROSS study. Conclusions: In highly selected EC and GEJC pts, definitive nCRT radiation doses do not appear to increase POC rates. Thus, 50 Gy can likely be delivered without increasing toxicity while also achieving a definitive dose for pts not able or willing to undergo subsequent surgery. [Table: see text]

Neural network dose models for knowledge-based planning in pancreatic SBRT

PURPOSE

Stereotactic body radiation therapy (SBRT) for pancreatic cancer requires a skillful approach to deliver ablative doses to the tumor while limiting dose to the highly sensitive duodenum, stomach, and small bowel. Here, we develop knowledge-based artificial neural network dose models (ANN-DMs) to predict dose distributions that would be approved by experienced physicians.

METHODS

Arc-based SBRT treatment plans for 43 pancreatic cancer patients were planned, delivering 30-33 Gy in five fractions. Treatments were overseen by one of two physicians with individual treatment approaches, with variations in prescribed dose, target volume delineation, and primary organs at risk. Using dose distributions calculated by a commercial treatment planning system (TPS), physician-approved treatment plans were used to train ANN-DMs that could predict physician-approved dose distributions based on a set of geometric parameters (vary from voxel to voxel) and plan parameters (constant across all voxels for a given patient). Patient datasets were randomly allocated, with two-thirds used for training, and one-third used for validation. Differences between TPS and ANN-DM dose distributions were used to evaluate model performance. ANN-DM design, including neural network structure and parameter choices, was evaluated to optimize dose model performance.

RESULTS

Remarkable improvements in ANN-DM accuracy (i.e., from > 30% to < 5% mean absolute dose error, relative to the prescribed dose) were achieved by training separate dose models for the treatment style of each physician. Increased neural network complexity (i.e., more layers, more neurons per layer) did not improve dose model accuracy. Mean dose errors were less than 5% at all distances from the PTV, and mean absolute dose errors were on the order of 5%, but no more than 10%. Dose-volume histogram errors (in cm³ ) demonstrated good model performance above 25 Gy, but much larger errors were seen at lower doses.

CONCLUSIONS

ANN-DM dose distributions showed excellent overall agreement with TPS dose distributions, and accuracy was substantially improved when each physician's treatment approach was taken into account by training their own dedicated models. In this manner, one could feasibly train ANN-DMs that could predict the dose distribution desired by a given physician for a given treatment site.

Intracoronary formation and retention of micro aggregates of leukocytes and platelets contribute to postischemic myocardial dysfunction

OBJECTIVE

Cardiac pump function and coronary regulation can be impaired after short-term ischemia. Recent studies with platelets (P) and neutrophils (PMN) yielded contradicting results about the "cellular" contribution to reperfusion injury.

METHODS

Isolated guinea pig hearts performing pressure-volume work were employed, external heart work (EHW), aortic flow (AF), coronary flow (CF) and heart rate (HR) serving as parameters of cardiac function. After global ischemia, human blood cells were given as bolus (1 min) during reperfusion (intracoronary hematocrit 7%). Expression of specific adhesion molecules (P: CD62P, CD41; PMN: integrin CD11b) was measured on cells before and after coronary passage (FACS analysis).

RESULTS

Postischemic recovery of pump function was significantly reduced in hearts with blood cell application (EHW: -cells 54 +/- 14%, +cells 41 +/-12%, p <0.05). Coronary response to bradykinin and reactive hyperemia were not effected. The blood-cell dependent functional loss was partly reduced by blocking CD18 (anti-CD 18) and completely abrogated by blockage of CD41 (lamifiban). The expression of CD11b on PMN and monocytes (M) and CD62P on platelets was significantly reduced in the coronary effluent and a significant decrease of CD41 on leukocytes occurred during coronary passage after ischemia. Increases in CD41 on PMN in the presence of lamifiban demasked intracoronary formation of micro aggregates (P/PMN). These micro aggregates were visualized by light microscopy. Electron microscopy revealed no significant microvascular plugging.

CONCLUSION

1) A specifically blood-cell induced loss of myocardial pump function has been demonstrated after short-term ischemia. 2) CD41 (= GpIIbIIIa) on P is responsible for this cardiac reperfusion damage. 3) The effect is causally linked to the formation of micro aggregates between PMN and P, but seems attenuated in the presence of erythrocytes as compared to effects reported from experiments in which PMN and P were applied singly or co-perfused. 4) Intracoronary retention of PMN, M and platelet-leukocyte micro aggregates seems to be transient, as adherence was not confirmed by electron microscopy.

[Changes and complication rate in surgery for thoracic esophageal carcinoma]

AIM

In an analysis over 22 years it was investigated which parameters have changed in the operative treatment of thoracic esophageal carcinoma over time and in how far they have influenced complication rate.

PATIENTS AND METHODS

Between 1978 and 1999 386 patients (350 men, 36 women) underwent resection for thoracic esophageal carcinoma (squamous cell carcinoma n=300, adenocarcinoma n=86). Cervical tumors were excluded from analysis. The time periods from 1978 to 1988 (n=242) and from 1989 to 1999 (n=144) were separately analyzed and compared with respect to age, sex, histological type, main tumor location, neoadjuvant therapy, method of operation, esophageal substitute and positioning of the substitute, R-status, pT/pN classification, UICC stage, number of dissected lymph nodes, complication rate, postoperative mortality and survival.

RESULTS

Comparison of the two time periods showed a significant increase in adenocarcinomas and main tumor location in the lower thoracic third of the esophagus. Furthermore, significant changes concerning the indication of neoadjuvant chemoradiation, operative approach, esophageal substitute, R-status and number of dissected lymph nodes were observed. Tumor stage (pT/pN classification and UICC stage) significantly shifted towards earlier stages. Total complication rate dropped tendentially form 68.5 % to 59.0 % (p=0.061). Hospital mortality was significantly reduced from 24 % to 12.5 %, whereas anastomotic leakages and multiorgan failure remained on a constant level. Median survival of R0 resected patients was significantly prolonged from 19 months to 34 months.

CONCLUSIONS

The increase of esophageal adenocarcinoma, a more strict patient selection (staging, functional status), standardization of operative technique as well as an optimized intensive care management are among the important changes in the operative management of thoracic esophageal carcinoma that have resulted in an improvement of prognosis of curatively resected patients. In spite of a more aggressive operative approach, i. e. lymph node dissection, operative mortality could be reduced by nearly 50 % in the face of a tendentially declining total complication rate.

Cutaneous eccrine glands of the foot pads of the rock hyrax (Procavia capensis, Hyracoidea, Mammalia)

In order to find correlations between skin gland morphology and specific ethological features, the cutaneous glands of the foot pads of Procavia capensis were studied by histological and various histochemical methods and by electron microscopy. In the foot pads, abundant specific eccrine skin glands occur, which consist of coiled tubular secretory portions and coiled ducts. The wall of the secretory part is composed of cuboidal glandular cells and myoepithelial cells. Among the glandular cells two types occur: clear and dark cells. Clear cells have numerous mitochondria and form a basal labyrinth, indicating fluid transport. Dark cells, which stain strongly with periodic acid-Schiff, contain a highly developed perinuclear Golgi apparatus, large amounts of rough endoplasmic reticulum and many secretory granules indicating production of glycoproteins. Cytokeratin (CK) 19 was found in secretory compartments and ducts, CK14 only in duct cells. Single cells of the secretory coils and ducts may be stained with antibodies against antimicrobial peptides. Some glandular cells contain proliferating cell nuclear antigen-positive nuclei especially in the ducts indicating an increased cell proliferation. Terminal transferase (TdT)-mediated d-UTP nick-end labeling-positive nuclei can be detected predominantly in the secretory coils and rarely in the transitional portions between ducts and end pieces. We suppose that proliferating cells migrate from the ducts to the secretory coils. The secretory product of the eccrine cutaneous glands seems to improve the traction between the foot pads of these animals and the steep and smooth rock formations among which they live.

Pneumococcal immunizations at flu clinics: the impact of community-wide outreach

This study examined the effectiveness of a community-wide outreach campaign to promote the use of pneumococcal vaccine at public flu immunization clinics, and assessed whether this intervention was more effective than simply making pneumococcal vaccination available at such clinics. In 1997, a community-wide outreach campaign promoting pneumococcal and influenza immunizations was launched in a 17 zip code area of Dutchess County, NY. The campaign was aimed at 7,961 Medicare beneficiaries urging them to obtain pneumococcal immunization from local flu clinics. Medicare reimbursement data were used to assess the countywide pneumococcal vaccination rate, and to analyze differences between rates for beneficiaries in the target area and elsewhere in the county. Between 1996 and 1997 there was a 94% increase in pneumococcal vaccination billed to Medicare beneficiaries in Dutchess County. The 1997 annual rate of pneumococcal immunization in the target area reached 16.3% versus 12.2% elsewhere in the county (p < 0.001), with an increase over the previous year of 8.7% and 5.6%, respectively. Nearly all of the increase is accounted for by pneumococcal vaccination delivered at flu clinics. It is possible to significantly increase the use of pneumococcal immunization by linking its delivery to community-based flu clinics and by developing local outreach strategies. The outreach campaign has a significant additive effect over simply making PPV available at flu shot clinics. Additional community-wide outreach can further improve pneumococcal immunization utilization rates.

Interstitial buffer capacity influences Na+/H+ exchange kinetics and oxyntic cell pHi in intact frog gastric mucosa

We studied the influence of interstitial buffer capacity and CO2-HCO3- on oxyntic cell intracellular pH (pHi) in intact frog gastric mucosa. Oxyntic cells in stripped gastric mucosa of Rana esculenta were loaded with the pH-sensitive dye 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein, and pHi was assessed fluorometrically. In the presence of a constant serosal and luminal pH, oxyntic cell pHi was dependent on the serosal but not the luminal concentration of cell-impermeable buffer ions such as N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), tris(hydroxymethyl)aminomethane, N-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid, and 3-(N-morpholino)propanesulfonic acid. The stepwise increase in oxyntic cell pHi from 6.74 +/- 0.05 with 1 mM HEPES to 7.23 +/- 0.08 was almost completely inhibited by removal of serosal Na+ and by amiloride and dimethyl amiloride, suggesting that it was largely due to Na+/H+ exchange. Increasing the serosal concentration of a CO2-HCO3- buffer from 1% CO2-4 mM HCO3- to 10% CO2-40 mM HCO3- increased oxyntic cell pHi from 7.03 +/- 0.06 to 7.39 +/- 0.07. This CO2-HCO(3-)-dependent pHi increase was also Na+ and amiloride sensitive, but high HCO3- concentrations increased pHi even in the absence of Na+, K+, or Cl-, and in the presence of omeprazole, bafilomycin A1, or acetazolamide. We suggest that in intact frog gastric mucosa, cellular proton extrusion acidifies the interstitial pH immediately adjacent to the basolateral membrane of the oxyntic cells in the absence of a high interstitial buffer capacity.(ABSTRACT TRUNCATED AT 250 WORDS)

Estradiol and progesterone replacement regimens for the induction of endometrial receptivity

Initiation of pregnancy in premature ovarian failure patients by use of donated oocytes fertilized in vitro requires establishment of a normal endometrial environment. We compared administration of estradiol (E2) and progesterone (P) by polysiloxane vaginal rings versus oral micronized E2 and P vaginal suppositories in 10 such patients. Serum E2 levels were similar between groups and similar to normally-cycling controls. With vaginal administration of E2, a burst effect was noted, with marked elevation 1 hour after insertion. The pattern with oral administration was more consistent, although marked conversion to estrone occurred. The P cylinder and suppositories delivered similar levels, with diminution of P in some patients with the cylinder. Despite apparent limitations, endometrial histology was normal after each cycle; both groups achieved pregnancies. Administration of E2 and P by polysiloxane vaginal rings achieved hormonal levels similar to oral micronized E2 and P vaginal suppositories. Endometrial biopsies after the stimulated cycle were appropriately mature.

Modulating actions of estradiol on gonadotropin-releasing hormone-stimulated prolactin secretion in postmenopausal individuals

The temporal aspects of estrogen treatment on serum prolactin concentrations basally and in response to a 10 micrograms intravenous injection of gonadotropin-releasing hormone were assessed in eight postmenopausal women. The response of prolactin to gonadotropin-releasing hormone stimulation was compared with that of follicle-stimulating hormone and luteinizing hormone obtained simultaneously. Basal serum prolactin increased significantly (p less than 0.001) in response to estrogen treatment, and a positive correlation was found between the serum concentrations of estradiol and prolactin (r2 = 0.266; p = 0.0011). Gonadotropin-releasing hormone induced a significant increase in serum prolactin concentrations, which was greater after 5 to 10 days of estrogen treatment compared with that in the estrogen-depleted state (p = 0.031). No correlation was found between gonadotropin-releasing hormone-stimulated prolactin and luteinizing hormone or follicle-stimulating hormone release. These data demonstrate that estrogen treatment of previously hypoestrogenemic postmenopausal women potentiates gonadotropin-releasing hormone-stimulated prolactin release. Furthermore, our data suggest that the previously described paracrine interaction between the gonadotropins and lactotropes exerted by gonadotropin-releasing hormone does not appear to be mediated via increased gonadotropin release.

Actions of estradiol on discrete attributes of the luteinizing hormone pulse signal in man. Studies in postmenopausal women treated with pure estradiol

We assessed the time-dependent impact of estradiol on properties of the luteinizing hormone (LH) pulse signal in 12 hypoestrogenemic postmenopausal volunteers studied basally and after 1, 5, 10, and 30 d of estradiol delivery via an intravaginal Silastic ring. Computerized analysis of the plasma LH time series revealed a significant decrease in LH pulse frequency within 24 h of estrogen treatment, followed by a secondary increase (days 5 and 10), and then a sustained decline (day 30) in LH pulsatility. Estradiol also significantly suppressed incremental and maximal (but not fractional) LH pulse amplitudes in a biphasic manner. In contrast, LH peak duration was invariant until day 30 of estradiol replacement. These observations indicate that the well recognized biphasic actions of estradiol on mean serum LH concentrations can be modeled in relation to specific and time-dependent alterations in LH pulse frequency and amplitude.

Pituitary self-priming actions of gonadotropin-releasing hormone. Kinetics of estradiol's potentiating effects on gonadotropin-releasing hormone-facilitated luteinizing hormone and follicle-stimulating hormone release in healthy postmenopausal women

We examined the kinetically distinct characteristics of estradiol's effects upon pituitary luteinizing hormone (LH) and follicle-stimulating hormone (FSH) release in response to pulses of exogenous gonadotropin-releasing hormone (GnRH) in healthy postmenopausal individuals. The putative self-priming actions of GnRH on LH and FSH release were tested by intravenous injections of equal paired doses of GnRH (10 micrograms) before and after 1, 5, 10, and 30 d of pure estradiol-17 beta delivery via an intravaginal silastic ring. Self-priming actions of GnRH, as defined by heightened gonadotropin release in response to the second pulse of GnRH compared with the first, were completely absent in the hypoestrogenemic state. However, estradiol administration unmasked GnRH self-priming in a time-dependent fashion, with maximal expression after 5 and 10 d of steroid replacement, followed by attenuation by 30 d. Since estradiol's modulation of GnRH action was expressed differentially on LH and FSH release, we suggest that such facilitation of GnRH-stimulated pituitary LH and FSH release may provide an additional mechanism for dissociated secretion of gonadotropic hormones in health or disease.

Endogenous opiates participate in the regulation of pulsatile luteinizing hormone release in an unopposed estrogen milieu: studies in estrogen-replaced, gonadectomized patients with testicular feminization

Administration of opiate receptor antagonists augmented pulsatile LH release in six estrogen-treated, orchidectomized individuals with testicular feminization. In response to an opiate antagonist, LH pulse frequency increased from 3.0 +/- 0.55 (+/- SE) to 5.0 +/- 0.45 pulses/8 h (P = 0.034). Since these patients have inborn target tissue resistance to androgens, we infer that sustained androgen action is not obligatory to the emergence of endogenous inhibitory opiate tone in man. Rather, these observations document that the suppressive effects of opiates on gonadotropin secretion can be unmasked in the presence of an unopposed estrogen milieu.