Oncology News

Machine Learning Predicts Thyroid Nodules

According to an ahead-of-print article to be published in the December 2019 issue of the American Journal of Roentgenology, researchers have validated a first-of-its-kind machine learning-based model to evaluate immunohistochemical (IHC) characteristics in patients with suspected thyroid nodules, achieving “excellent performance” for individualized noninvasive prediction of the presence of cytokeratin 19, galectin 3, and thyroperoxidase, based on CT images.

“When IHC information is hidden on CT images,” principal investigator Jiabing Gu explains, “it may be possible to discern the relation between this information and radiomics by use of texture analysis.”

To assess whether texture analysis could be utilized to predict IHC characteristics of suspected thyroid nodules, Gu and colleagues from China’s University of Jinan enrolled 103 patients—with a training cohort-to-validation cohort ratio of approximately 3:1—with suspected thyroid nodules who had undergone thyroidectomy and IHC analysis from January 2013 to January 2016. All 103 patients—28 men and 75 women with a median age of 58 years and a range of 33 to 70 years—underwent CT before surgery, and 3D Slicer v 4.8.1 was used to analyze images of the surgical specimen.

To facilitate test-retest methods, 20 patients were imaged in two sets of CT series within 10 to 15 minutes, using the same scanner and protocols, without contrast administration. These images were used only to select reproducible and nonredundant features, not to establish or verify the radiomic model.

The Kruskal-Wallis test (SPSS v 19, IBM) was employed to improve classification performance between texture feature and IHC characteristic. Gu and colleagues considered characteristics with p<0.05 significant. The feature-based model was trained via support vector machine methods and assessed with respect to accuracy, sensitivity, specificity, and other measures including independent validation. From 828 total features, 86 reproducible and nonredundant features were selected to build the model.

The best performance of the cytokeratin 19 radiomic model yielded accuracy of 84.4% in the training cohort and 80% in the validation cohort. Meanwhile, the thyroperoxidase and galectin 3 predictive models showed accuracies of 81.4% and 82.5% in the training cohort and 84.2% and 85% in the validation cohort, respectively.

Noting that cytokeratin 19 and galectin 3 levels are high in papillary carcinoma, Gu says that these models can help radiologists and oncologists identify papillary thyroid cancers, “which is beneficial for diagnosing papillary thyroid cancers earlier and choosing treatment options in a timely manner.”

Ultimately, Gu says, “this model may be used to identify benign and malignant thyroid nodules.”

— Source: American Roentgen Ray Society


Postoperative Dose Escalation Not Necessary for HNSCC

Patients with head and neck squamous cell carcinoma (HNSCC) may not benefit from high-dose postoperative radiation as compared with standard dosing, according to new research recently published in Head & Neck. Currently, adjuvant dosing of radiation therapy is often at the treating physician’s discretion, with doses ranging from as low as 57.6 Gy to as great as 70 Gy, according to Thomas J. Galloway, MD, an associate professor in the department of radiation oncology at Fox Chase Cancer Center in Philadelphia.

“It is sometimes human nature to think that more is better, and the guidelines on the topic have a range of doses with no specific recommendation,” Galloway says. “However, according to this data, there is no evidence to support routine dose escalation in the adjuvant radiation of head and neck cancer patients.”

In order to see whether a specific group of patients with HNSCC benefited more from high-dose radiation than others, Galloway and colleagues queried the National Cancer Database for patients with nonmetastatic HNSCC diagnosed between 2004 and 2013 who underwent adjuvant radiation. Standard-dose radiation was defined as an equivalent dose in 2 Gy of ≥56.64 Gy and ≤60 Gy, and high-dose radiation was an equivalent dose in 2 Gy of >60 Gy and <70 Gy. Patients with equivalent doses below and above the designated range were thought to have either inadequate or salvage therapy, respectively.

The study included 15,836 patients who received either standard-dose or high-dose radiation. Of those who received high-dose radiation, 14.3% received a dose >66 Gy. Undergoing high-dose radiation therapy was associated with a 9% increased risk for mortality compared with standard-dose radiation. Among a high-risk group of 3,291 patients with nonoropharynx or known human papilloma virus-negative oropharynx disease with positive margins, five or more positive lymph nodes, and/or extranodal extension, treatment with high-dose radiation conferred no survival advantage.

“Our study showed that no matter how we looked at the data, we could see no proof that survival was better with increased radiation,” Galloway says.

This finding demonstrates two things, according to Galloway. First, when choosing between two radiation doses that lead to similar outcomes, logic dictates selection of the lower dose to cause fewer side effects to the patient. Second, these data show that there is still an unclear picture of the best dose of radiation therapy to deliver after HNSCC surgery.

“This is something we should be looking into,” Galloway says. “It would be very intriguing to find out if there is a way to determine the best dose for each individual patient.”

Source: Fox Chase Cancer Center


Multidisciplinary Tumor Board Coordination Increases Head and Neck Cancer Survival

Patients with head and neck cancer lived longer when their treatment was coordinated by a dedicated multidisciplinary tumor board, according to a study by clinicians at Fox Chase Cancer Center at Temple University Hospital and the departments of radiation oncology and otolaryngology-head and neck surgery at the Lewis Katz School of Medicine at Temple University in Philadelphia. The study was recently published in The Laryngoscope.

“For decades, everyone has understood that a multidisciplinary group of doctors and staff caring for patients with cancer has been best practice, but one thing was not clear: if the added time and effort of this multidisciplinary approach improved patient outcomes in terms of survival,” says Jeffrey C. Liu, MD, FACS, of the Lewis Katz School of Medicine and Fox Chase Cancer Center. “We show here that there is a measurable difference in survival after implementation of a multidisciplinary tumor board.”

Liu and his team had an opportunity to measure the effects of coordination of care by a tumor board when a dedicated head and neck surgeon joined the staff of Temple University Hospital in 2011 and instituted a formalized multidisciplinary tumor board. From 2006 to 2011, patients with head and neck cancer had treatment teams composed of otolaryngologists, medical oncologists, and radiation oncologists who met to discuss patient cases, but meetings were less formalized. Starting in 2011, the head and neck surgeon expanded the tumor board to include not only medical oncologists and radiation oncologists, but also neuroradiology, speech therapy, nutrition, pathology, dental services, and social work. Meeting frequency was standardized to weekly conferences, with new and existing cases presented for decision making and discussion. 

Liu and colleagues reviewed outcomes from 224 patients with head and neck cancer treated at Temple University Hospital from 2006 to 2015; 98 patients were treated prior to the formalization of the tumor board and 126 were treated after the formalization. Patients treated after the establishment of the formal multidisciplinary tumor board had better overall and disease-specific survival. These patients had a 52% decreased risk for death. According to Liu, these data reflect that expanded multidisciplinary tumor boards play a vital role in the coordination of head and neck cancer care.

“Cancer treatment can be a lengthy process,” Liu says. “It is not unheard of for a treatment plan to take three months to execute. The tumor board helps to make sure that all of the pieces are in place to execute appropriate treatment as quickly as possible.”

Source: Fox Chase Cancer Center