Could machine learning improve the prediction of pelvic nodal status of prostate cancer patients? Preliminary results of a pilot study.

We tested and compared performances of Roach formula, Partin tables and of three Machine Learning (ML) based algorithms based on decision trees in identifying N+ prostate cancer (PC). 1,555 cN0 and 50 cN+ PC were analyzed. Results were also verified on an independent population of 204 operated cN0 patients, with a known pN status (187 pN0, 17 pN1 patients). ML performed better, also when tested on the surgical population, with accuracy, specificity, and sensitivity ranging between 48-86%, 35-91%, and 17-79%, respectively. ML potentially allows better prediction of the nodal status of PC, potentially allowing a better tailoring of pelvic irradiation.

Late rectal bleeding after 3D-CRT for prostate cancer: development of a neural-network-based predictive model.

The aim of this study was to develop a model exploiting artificial neural networks (ANNs) to correlate dosimetric and clinical variables with late rectal bleeding in prostate cancer patients undergoing radical radiotherapy and to compare the ANN results with those of a standard logistic regression (LR) analysis. 718 men included in the AIROPROS 0102 trial were analyzed. This multicenter protocol was characterized by the prospective evaluation of rectal toxicity, with a minimum follow-up of 36 months. Radiotherapy doses were between 70 and 80 Gy. Information was recorded for comorbidity, previous abdominal surgery, use of drugs and hormonal therapy. For each patient, a rectal dose-volume histogram (DVH) of the whole treatment was recorded and the equivalent uniform dose (EUD) evaluated as an effective descriptor of the whole DVH. Late rectal bleeding of grade ≥ 2 was considered to define positive events in this study (52 of 718 patients). The overall population was split into training and verification sets, both of which were involved in model instruction, and a test set, used to evaluate the predictive power of the model with independent data. Fourfold cross-validation was also used to provide realistic results for the full dataset. The LR was performed on the same data. Five variables were selected to predict late rectal bleeding: EUD, abdominal surgery, presence of hemorrhoids, use of anticoagulants and androgen deprivation. Following a receiver operating characteristic analysis of the independent test set, the areas under the curves (AUCs) were 0.704 and 0.655 for ANN and LR, respectively. When evaluated with cross-validation, the AUC was 0.714 for ANN and 0.636 for LR, which differed at a significance level of p = 0.03. When a practical discrimination threshold was selected, ANN could classify data with sensitivity and specificity both equal to 68.0%, whereas these values were 61.5% for LR. These data provide reasonable evidence that results obtained with ANNs are superior to those achieved with LR when predicting late radiotherapy-related rectal bleeding. The future introduction of patient-related personal characteristics, such as gene expression profiles, might improve the predictive power of statistical classifiers. More refined morphological aspects of the dose distribution, such as dose surface mapping, might also enhance the overall performance of ANN-based predictive models.

Interphysician variability in defining the planning target volume in the irradiation of prostate and seminal vesicles.

With the improvements in treatment accuracy the risk exists of over-reliance on the capability of the physician to estimate the extent of the tumour. We assessed the variability between six radiotherapists in defining the planning target volume (PTV) on CT slices for three prostate tumour cases. Percentage differences between measured volumes and mean values calculated for each case ranged from -53.64 to +60.48% (SD 36.00%). There is a considerable variation in delineating the PTV, both in the cranio-caudal direction and in the in-slice extension of the areas drawn on each slice (standard deviations ranged from 0.35 to 2.64 cm2). We also checked the uncertainty in the shape and position of the contours on each CT image. The analysis was performed on three slices of one test case. As we expected, the uncertainty seems largest for seminal vesicle slices and smallest for prostatic apex slices. These results endorse the need for uncertainty analysis of all departmental processes in order to define a detailed protocol and consequently to minimize the interphysician differences in PTV delineations.

Continuous infusion of carboplatin during conventional radiotherapy treatment in advanced squamous carcinoma of the cervix uteri IIB-IIIB (UICC). A phase I/II and pharmacokinetic study.

OBJECTIVE: A prospective, single-arm phase-I/II trial performed to assess the efficacy and toxicity of the concomitant use continuous infusion of low-dose carboplatin and pelvic conventional radiotherapy in patients with locally advanced squamous cell carcinoma of the cervix. MATERIALS AND METHODS: Between January and July 1994, a total of 12 patients consecutively diagnosed to have squamous cell carcinoma of the cervix uteri stages IIB-IIIB UICC-TNM (five patients, IIB; and seven patients, IIIB) entered the study. All patients were evaluated by a gynecologist and a radiation oncologist and were submitted to standard pretreatment staging procedures. Radiation was delivered with 10-MeV photon beams with the shrinking-field technique. The patients received 2 Gy radiotherapy daily, 5 fractions per week, up to a planned total of 60 Gy in 6 weeks to the primary tumor and 46 Gy in 4 weeks to the whole pelvis. Irradiation was performed using four fixed orthogonal fields. One intracavitary insertion, 8 Gy to point A (dose rate, 1.1 Gy/h), was performed immediately after external pelvic irradiation. Carboplatin (12 mg/m2/day) was also administered in a continuous infusion, starting 1 day before the first fraction of radiotherapy. The platinum in plasma and urine, as well as the platinum concentration in the cytosols of lymphocytes and tumor, was measured weekly. RESULTS: A complete response was seen in nine (75%) of the 12 patients. Of the nine patients who achieved a complete remission, only one had subsequent failure in the pelvis. The total pelvic failure rate was 33.3% (four of 12 patients). With a median follow-up time of 20 months, the actuarial survival rate at 24 months was 64.8%. All patients completed the treatment without major protocol violations. Grade-2 leukopenia (in nine patients) and grade-1 nausea and vomiting (in five) were the most common acute toxicities. There was one grade-3 hematologic toxicity. Grade-3 late complications were observed in 16.6% of cases (two of 12 patients). On days 28 and 42 of the treatment, the mean total platinum plasma concentrations were 491 micrograms/L (SD = 129) and 672 micrograms/L (SD = 160), and the ultrafilterable fraction was 8-10%. At the same time points, the concentration in lymphocytes was constant at 21 picograms (pg) platinum/lymphocyte. The levels of platinum concentration measured on days 14 and 28 in the cytosols of tumor cells were 0.3 microgram/g (SD = 0.1) and 0.93 microgram/g (SD = 0.2). CONCLUSION: The combination of continuous infusion of carboplatin and radiotherapy at the aforementioned doses in patients with locally advanced cervical carcinoma resulted in a relatively low frequency of significant acute and late complications. Platinum in normal tissue (picograms per lymphocyte) was stable from week 1 of treatment, whereas the platinum steady state in plasma and in tumor cells was not reached in 6 weeks and was below that required in vitro to produce radiopotentiation. Further studies to determine the optimal dose of carboplatin and irradiation are needed prior to the initiation of phase-III studies.

[Locally advanced carcinoma of the cervix uteri (stage IIB-IIIB TNM-UICC): radiotherapy combined with simultaneous daily low-dose platinum. Phase II study]. / Carcinoma della cervice uterina localmente avanzato (stadio IIB-IIIB TNM-UICC): radioterapia in associazione a basse dosi concomitanti e giornaliere di platino. Studio della fase II.

A prospective, single arm, phase-II trial was performed to assess the efficacy and local toxicity of the combination of low doses of platin and pelvic radiotherapy in patients with locally advanced carcinoma of the cervix. January, 1993, through August, 1994, twenty-three previously untreated patients with squamous carcinoma (stages IIB-IIIB UICC) entered the study. All patients were examined by a gynecologist and by a radiation oncologist and then submitted to conventional pretreatment staging procedures. Nine patients were classified as stage IIB and 14 patients as stage IIIB. Radiotherapy consisted of 60 Gy external beam irradiation (46 Gy to pelvis + 14 Gy boost to cervix uteri and parametria) plus one low dose rate intracavitary treatment to a dose of 8 Gy to point A. Cisplatin (3 mg/m2/day) or carboplatin (12 mg/m2/day) was also given for 6 weeks starting on radiotherapy day 1. The treatment was well tolerated and no patient required radiotherapy discontinuation. With a median follow-up time of 20 months, complete response was seen in 74% (17/23) of the patients. One of the 17 patients who achieved a complete remission, during follow-up, relapsed in the pelvis and one developed lung metastases. Total failure rate in the pelvis was 30.5% (7/23). Distant metastases were observed in 17.5% (4/23) of the patients. Actuarial overall and disease-free survival rates at 33 months were 69.1% and 65.2%, respectively. Late gastrointestinal toxicity (grade 3) occurred in 8.6% (2/23) of patients, with one patient developing a rectal ulcer-which was submitted to colostomy- and one patient a vaginal necrosis. The combination of platin and radiotherapy appears to be an effective regimen for the patients with locally advanced carcinoma of the cervix and caused a relatively low rate of late gastrointestinal complications.

[The irradiated volume in stage III non-small-cell carcinoma of the lung. Comparison of outdated and new techniques]. / Il volume irradiato nel carcinoma non-microcitoma del polmone allo stadio III. Confronto tra vecchie e nuove metodiche.

March 1988 through April 1992, three hundred and ninety-six patients affected with bronchogenic carcinoma were treated at the Department of Radiation Oncology of the S. Anna Hospital, Como, Italy. A hundred and ten patients presenting stage-III non-small-cell lung carcinoma were evaluable. All evaluable patients underwent radiation therapy alone, with either palliative or curative purposes. Two main periods can be distinguished: in period A, before June 30th, 1990, treatment planning included conventional techniques, with no simulators; the patients were treated with opposing anteroposterior fields only. In period B, after July 1st, 1990, either the simulator alone was used or a simulator, a CT unit and a treatment planning computer system were combined; anteroposterior opposing fields or multiportal technique were used. Median overall survival was 10 months. Independent of treatment goals, the irradiated volume was markedly different in the patients treated in period A than in those treated in period B. A marked and statistically significant increase in survival was observed in group B. Survival also increased in patients treated with doses > 40 Gy, but only if treatment planning had used adequate technology and accuracy. To conclude, better survival can be achieved only by improving treatment accuracy and quality.