A Cost-Effective, Machine Learning-Based New Unified Risk-Classification Score (NU-CATS) for Patients with Endometrial Cancer

PURPOSE/OBJECTIVE(S)

Treatment for endometrial cancer (EC) with radiotherapy is increasingly guided by molecular risk classifications. Derived from genomic profiling of The Cancer Genome Atlas (TCGA) project, several EC risk classification systems, including ProMisE and Leiden/TransPORTEC, have been developed. However, the current systems were developed on a relatively homogeneous population. Black or African American (BOAA) patients have consistently been demonstrated to have worse stage-adjusted prognosis than Caucasians. Given this, we intended to develop a new unified risk classification system (NU-CATS) for EC patients using machine learning (ML) utilizing datasets with demographically diverse populations.

MATERIALS/METHODS

TCGA-Uterine Corpus Endometrial Carcinoma (n = 596), Memorial Sloan Kettering-Metastatic Events and Tropisms (MSK-MET, n = 1,315) and the American Association for Cancer Research Project Genomics Evidence Neoplasia Information Exchange (AACR-GENIE, n = 4,561) were used to identify genetic alterations and clinicopathological features, including age, race, stage, histologic grade and features, and distribution of metastatic disease. Software packages including Keras, Pytorch, and Scikit Learn were tested to build artificial neural networks (ANNs) with a binary output as either intra-abdominal metastatic lesions vs. non-metastatic. A 5-layered ANN (5-6-4-2-1) using 5 inputs ('age at surgery', 'histology', 'race', 'mismatch repair status' and 'TP53'). The optimal performing ANN was selected and cross validated. The weights and biases of the trained ANN were used to reconstruct the algorithm.

RESULTS

BOAA patients with EC have worse prognosis than Caucasians, adjusting for TP53 or POLE mutation status. TP53 is the most common gene differentially altered by race in EC. Over 75% of BOAA patients carry TP53 mutations as compared to approximately 40% of Caucasians. Older age is associated with an increasing likelihood of TP53 mutations, high risk histology, and distant metastasis. For patients above age 70, 91% of BOAA and 60% of Caucasian EC patients carry TP53 mutations. The NU-CATS that incorporates age, race, histology, mismatch repair (MMR) status, and TP53 mutation status showed 75% accuracy in prognosticating intra-abdominal metastasis. A higher NU-CATS (>50) is associated with about 2-fold increased risk of having positive pelvic or para-aortic lymph nodes (LNs) and distant. NU-CATS was shown to outperformed TransPORTEC model for estimating risk of FIGO Stage I/II disease progression and survival in BOAA EC patients.

CONCLUSION

Despite adjusting for molecular classification, race and age retain prognostic importance in EC. NU-CATS, a ML-based, cost-effective algorithm, incorporates diverse clinicopathologic and molecular variables of EC, and yields superior prognostication of the risk of nodal involvement, distant metastasis, disease progression, and overall survival as compared to other classification systems.

An Investigation of High-Z Material for Bolus in Electron Beam Therapy

PURPOSE/OBJECTIVE(S)

Electron beams are frequently used for superficial tumors including: primary skin cancers, surgical scars, cutaneous lymphomas and benign conditions like keloids. However, due to electron beam characteristics the surface dose is 75-95% of the prescribed dose depending on beam energy thus requiring placement of bolus to augment surface dose. Various types of boluses; wet gauge/towel, superflab, superstuff, aquaplast, 3D printed plastics, customized dot-decimal bolus, thermoshield, brass and tantalum wire mesh are commonly used in clinics with variable difficulties that are not perfect. As majority of these devices do not snug to the skin contour and create air-gap that is known to produce significant dose perturbations creating hot and cold spot. The variable dosimetry has cosmetic implications and potentially increased risk of recurrence. A new cloth-like high-Z materials; Tungsten, (Z = 74) and Bismuth, (Z = 83) impregnated with silicone gel are now commercially available which is investigated in this study for bolus purpose.

MATERIALS/METHODS

Commercially available, super soft silicone-gel based submillimeter thin Tungsten and Bismuth sheets were investigated for bolus properties in electron beams in the range of 6-12 MeV using 10x10 cm2 applicator. These materials were tested on various body contour for perfect snug without any air traps. Using parallel plate ion chamber measurements were performed in a solid water phantom on a medical linear accelerator machine. Depth dose characteristics were measured to optimize the thickness for surface dose to be 100% for selected electron therapy.

RESULTS

Surface dose for a 10x10 cm2 cone for 6, 9, 12 MeV beams were measured to be 75.1%, 80.1% and 86.2%, respectively. Silicone-gel Tungsten and Bismuth sheets produce significant electrons thus increasing surface dose. Based on measured depth dose, our data showed that Tungsten sheets of 0.14 mm, 0.18 mm and 0.2 mm and Bismuth sheets of 0.42 mm, 0.18 mm and 0.2 mm provide 100% surface dose for 6, 9 and 12 MeV beams, respectively without any changes in depth dose except increasing surface dose.

CONCLUSION

The new Tungsten and Bismuth based silicone-gel clothlike sheets are extremely soft but high tensile metallic bolus materials that can fit flawlessly on any skin contour without any air trap or variable thickness or spillage of the water as in conventionally used wet gauge. Only 0.2 mm thick sheets are needed for 100% surface dose without degradation of the depth dose characteristics. These materials are reusable and ideal for bolus in electron beam treatment. This investigation opens a new frontier in designing new bolus materials optimum for patient treatment.

Treatment of Mantle and Inverted Y Fields in Modern Era: A VMAT Approach

PURPOSE/OBJECTIVE(S)

Hodgkin's lymphoma has been treated with combined modality therapy (chemotherapy and radiation) with a very high degree of success. Total Nodal Irradiation (TNI) performed with large AP/PA mantle fields for treatment of axillary, cervical, and mediastinal lymphatics, provide adequate coverage to the mediastinum and bilateral axillae and hila, while blocking lungs. The para-aortic and pelvic lymph nodes are treated with the so called inverted-Y AP/PA fields, which often includes the spleen in cases of TNI. Multileaf Collimators (MLC) have been tried, but due to the irregular shape of the fields and necessity of island blocking in 3D treatment, they have not been successful in full elimination of Cerrobend blocks. We hypothesize that using two or three matched Volumetric Modulated Arc Therapy (VMAT) fields will not only eliminate a need for Cerrobend blocks or island blocks, but will also provide better target coverage and better organs at risk (OAR) sparing.

MATERIALS/METHODS

Under IRB study, 10 patients were retrospectively planned using two or three matched VMAT technique for mantle and inverted-Y treatments of TNI that had been previously treated using MLC and Cerrobend block combination. Pinnacle treatment planning system version 16.2.1 was used to generate plans using mantle/inverted-Y technique and corresponding VMAT plans using 2-3 arcs per isocenter (2 isocenters per plan). Optimization was performed to cover targets with the prescribed dose of 1500 cGy in 10 fractions per institutional protocol. The VMAT plans were compared with traditional 3D plans.

RESULTS

VMAT consistently provided better or similar results to traditional field arrangements. Target coverage: V15Gy - 95.45% vs 77.99% (p = 0.00017), OAR coverage: total lung V5Gy 63.7% vs 68.8% (p = 0.016), bone marrow mean dose 539.1 cGy vs 727.8 cGy (p = 0.00047), Integral Dose 464.1 mJ vs 573.9 mJ (p = 0.0026). Low isodose lines- mean volume of 5 Gy isodose line was not significantly different - 24036 cc vs 25091 cc (p = 0.271). Cord maximum dose was 40% lower for VMAT plans (p = 0.00006). Mean bladder dose was similar in VMAT plans compared to 3D plan - 821.7 cGy vs 804.9 cGy (p = 0.454). One counter-intuitive result is that the mean integral dose for 10 patients was 24% lower for VMAT plans.

CONCLUSION

VMAT based mantle fields for TNI eliminates Cerrobend blocks and improve dosimetry significantly for target volumes and all OARs; including bone marrow, which plays important role in patient's recovery after chemotherapy, radiation and often stem cell transplantation in recurrent disease. Lower integral dose for VMAT plans is explained by the large irradiated in-fields and small out-of-field volumes. The VMAT process requires minimal effort for optimization and is economical compared to the traditional planning, while improving the target coverage and decreasing dose to OARs.

A multi-institutional analysis of intraoperative radiotherapy for early breast cancer: Does age matter?

BACKGROUND

Single-session intraoperative radiation therapy (IORT) minimizes treatment demands associated with traditional whole breast radiation therapy (WBRT) but outcomes on local disease control and morbidity among the elderly is limited.

METHODS

A multi-institutional retrospective registry was established from 19 centers utilizing IORT from 2007 to 2013. Patient, tumor, and treatment variables were analyzed for ages <70 and ≥70.

RESULTS

We evaluated 686 patients (<70 = 424; ≥70 = 262) who were margin and lymph node negative. Patients <70 were more likely to have longer operative time, oncoplastic closure, higher rates of IORT used as planned boost, and receive chemotherapy and post-operative WBRT. Wound complication rates were low and not significantly different between age groups. Median follow-up was 1.06 (range 0.51-1.9) years for < 70 and 1.01 (range 0.5-1.68) years for ≥ 70. There were 5 (0.73%) breast recurrences (4 in <70 and 1 ≥ 70, p = 0.65) and no axillary recurrences during follow-up.

CONCLUSIONS

IORT was associated with a low rate of wound complication and local recurrence on short-term follow-up in this cohort.

Reaction of Field-grown Sericea Lespedeza to Selected Meloidogyne spp

Five sericea lespedeza (Lespedeza cuneata [Dumont] G. Don) breeding lines resistant to Meloidogyne incognita, M. incognita acrita, and M. hapla as seedlings in greenhouse tests and two varieties were resistant to M. incognita acrita in field experiments. Root-knot galling and larvae numbers were less for resistant entries than for the susceptible check when grown in root-knot infested field soil for three growing seasons. Forage yields were as much as 57 times greater for resistant entries than for susceptible check. Lines resistant to M. incognita acrita generally appeared more tolerant to M. javanica than the susceptible check in the field. A population predominantly M. incognita acrita built up on cotton (Gossypium hirsutum L.) and hairy vetch (Vicia villosa L.) shifted to M. javanica under sericea lespedeza.