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.

Salvage radiotherapy in patients affected by oligorecurrent pelvic nodal prostate cancer.

PURPOSE: Metastasis-directed therapy (MDT) is an investigational treatment option in patients with oligorecurrent prostate cancer (PCa). The aim of this retrospective study is to report oncologic outcome and toxicity of elective nodal radiotherapy (ENRT) in PCa patients affected by pelvic nodal oligorecurrence. METHODS: 41 consecutive patients were treated with salvage radiotherapy. At biochemical recurrence after primary treatment, oligorecurrent disease was detected by positron emission tomography (PET) in 94% of the patients. Image-guided intensity modulated radiation therapy (IMRT) was delivered using tomotherapy. 83% of the patients received androgen deprivation therapy (ADT) in combination with ENRT. Survival analysis was performed with Kaplan-Meier method, log-rank test was used to analyze associations between survival end-points and clinical parameters. Multivariate analysis was performed using Cox proportional hazards regression models. Toxicity was registered according to Common Terminology Criteria for Adverse Events (CTCAE) v4.0. RESULTS: The median at follow-up was 33.6 months. At 3 years, overall survival (OS), cancer-specific survival (CSS), and biochemical progression-free survival (b-PFS) were 89%, 92%, and 53%, respectively. At univariate analysis, all survival end-points were correlated with the number of positive pelvic lymph nodes at oligorecurrence (≤ 3 vs > 3). Biochemical-PFS was correlated with PSA (p = 0.034) and PSA doubling time (p = 0.004) at oligorecurrence. At multivariate analysis, no independent variable was statistically significant. No patient experienced grade ≥ 2 late toxicity after radiotherapy. CONCLUSIONS: The number of metastatic lymph nodes and PSA doubling time seems to be important prognostic factors in the pelvic oligorecurrent setting. Salvage radiotherapy combined with short-course ADT might be a valid treatment strategy.

Radiation-induced myelopathy in long-term surviving metastatic spinal cord compression patients after hypofractionated radiotherapy: a clinical and magnetic resonance imaging analysis.

BACKGROUND AND PURPOSE: Hypofractionated radiotherapy is often administered in metastatic spinal cord compression (MSCC), but no studies have been published on the incidence of radiation-induced myelopathy (RIM) in long-term surviving patients. Our report addresses this topic. PATIENTS AND METHODS: Of 465 consecutive MSCC patients submitted to radiotherapy between 1988 and 1997, 13 live patients (seven females, six males, median age 69 years, median follow-up 69 months) surviving for 2 years or more were retrospectively reviewed to evaluate RIM. All patients underwent radiotherapy. Eight patients underwent a short-course regimen of 8 Gy, with 7 days rest, and then another 8 Gy. Five patients underwent a split-course regimen of 5 Gy x 3, 4 days rest, and then 3 Gy x 5. Only one patient also underwent laminectomy. Full neurological examination and magnetic resonance imaging (MRI) were performed. RESULTS: Of 12 patients submitted to radiotherapy alone, 11 were ambulant (eight without support and three with support) with good bladder function. In nine of these 11 patients, MRI was negative; in one case MRI evidenced an in-field relapse 30 months after the end of radiotherapy, and in the other, two new MSCC foci outside the irradiated spine. In the remaining patient RIM was suspected at 18 months after radiotherapy when the patient became paraplegic and cystoplegic, and magnetic resonance images evidenced an ischemic injury in the irradiated area. The only patient treated with surgery plus postoperative radiotherapy worsened and remained paraparetic. Magnetic resonance images showed cord atrophy at the surgical level, explained as an ischemic necrosis due to surgery injury. CONCLUSIONS: On the grounds of our data regarding RIM in long-term surviving MSCC patients, we believe that a hypofractionated radiotherapy regimen can be used for the majority of patients. For a minority of patients, more protracted radiation regimens could be considered.

Short-course palliative radiotherapy in non-small-cell lung cancer: results of a prospective study.

From February 1993 to October 1997, 91 consecutive patients with inoperable (stage IIIB-IV) histologically confirmed non-small-cell lung cancer underwent palliative hypofractionated radiotherapy. Recently, the Medical Research Council studies on hypofractionated short-course radiotherapy (8.5 Gy x 2) have reported high control of symptoms caused by thoracic disease without toxicity. Based on these experiences and our previous positive trial on short-course radiotherapy (8 Gy x 2) in metastatic spinal cord compression, a prospective study of short-course palliative radiotherapy in non-small-cell lung cancer was carried out. The regimen was 16 Gy given in two 8-Gy fractions, 1 week apart. Eighty-one patients were evaluable for response to treatment. Forty-eight (59%) patients were 65 years or older. Forty (49%) patients were naive to radiotherapy, whereas 41 (51%) had previous cisplatin-based chemotherapy. All but four stage IV patients (95%) had poor Eastern Cooperative Oncology Group performance status (i.e., 2-3). Clinical palliation was achieved in 62 (77%) patients. Performance status improved in 59 (73%) patients. The median palliation time ranged from 28% to 57% of patient survival. The median survival from the beginning of treatment was 148 days (range, 5-681 days). No difference in overall survival according to stage and previous chemotherapy was observed. Only performance status conditioned survival (performance status 1-2 vs. performance status 3; p = 0.0289). Short-course radiotherapy gave good results in terms of clinical palliation for thoracic symptoms, even in patients with poor performance status and pretreated with chemotherapy. The median palliation time was approximately 50% of patient survival time. Treatment was generally well tolerated-only 4 (5%) patients experienced World Health Organization grade III dysphagia. No late toxicity was recorded. The two-fraction regimen had social and economic advantages compared with the conventional ones.

Adjuvant radiochemotherapy in high-risk rectal cancer results of a prospective non-randomized study.

AIMS AND BACKGROUND: In 1990 the National Institutes of Health Consensus Conference recommended adjuvant combined therapy for patients with radically resected rectal cancer at high risk for relapse (ie, stage II-III). The purpose of our prospective non-randomized study was to verify the feasibility and effectiveness of postoperative radiochemotherapy in terms of improvement in disease-free and overall survival in this patient subgroup. STUDY DESIGN: From January 1990 to October 1998, 191 consecutive patients with radically resected stage II-III rectal cancer were treated. A total of 159 patients with a 24-month follow-up were assessable for toxicity and survival. Anterior resection was performed in 129 (81%) and abdomino-perineal resection in 30 (19%) patients. Fifty-four (34%) stage II and 105 (66%) stage III patients entered the study. Within 45-60 days of surgery, all patients received 5-fluorouracil chemotherapy at the dose of 500 mg/m2 as an i.v. bolus on days 1-5, every 4 weeks, for 6 cycles. Chemotherapy cycles III and IV were administered at the same daily dose on radiotherapy days 1-3 and 29-31. Radiotherapy consisted of 45 Gy/25 fractions plus a boost dose of 5.4 Gy. RESULTS: After a median follow-up of 57 months (range, 25-123), overall recurrent disease was reported in 58 (36%) patients: local, systemic, and both local and systemic relapses in 12 (8%), 37 (23%) and 9 (6%) cases, respectively. According to local extension, recurrence rates were 15% and 48% in stage II and III, respectively. Five-year overall and disease-free survival were 71% and 66%, respectively. Overall survival was 87% in stage II and 62% in stage III patients, and disease-free survival was 84% and 56% in stage II and III disease, respectively. According to univariate and multivariate analyses, significant prognostic factors for better tumor control were: stage (II vs III, P <0.001), the number of involved nodes (< or = 3 vs > 3, P <0.0001), and no extracapsular node invasion (P <0.0001). The recommended dose of the combined radiochemotherapy regimen was generally well tolerated. The incidence of any > or = grade 3 acute toxicity (according to the WHO scale) was 13% diarrhea, 11% proctitis, 5% perineal dermatitis and 4% myelosuppression. Four (3%) patients had radiotherapy-related severe late toxicity which required surgery. CONCLUSIONS: The study provided recurrence rates and survival similar to other adjuvant radiochemotherapy regimens published in the literature. However, in view of the low 5-year survival rate recorded in stage III patients, a different approach should be investigated.

Carboplatin plus pemetrexed for platinum-pretreated, advanced non-small cell lung cancer: a retrospective study with pharmacogenetic evaluation.

PURPOSE: In the present study, the combination of carboplatin (CBDCA) plus pemetrexed (PMX) for the treatment of patients with platinum-pretreated, pemetrexed-naïve, advanced non-small cell lung cancer (NSCLC) was investigated. Also, single nucleotide polymorphisms (SNPs) at the XRCC3, XPD, ERCC1, GARFT, DHFR, and TS genes were investigated. METHODS: Eighty patients treated with CBDCA/PMX at two Italian institutions were evaluable. Of these, 73 patients had blood samples collected for pharmacogenetic evaluation. RESULTS: Overall, the median age was 59 years (26-78), 59 patients (73.7%) had a performance status of 0, and 34 patients (42.5%) had stage IIIB disease. Thirty-eight patients (47.5%) had responded to prior first-line platinum-based therapy. Study treatment resulted into one complete and 33 partial responses for an overall response rate of 42.5%. The disease control rate was 77.5%. The median progression-free survival (PFS) and overall survival (OS) were 5.8 and 17.4 months, respectively. Responders achieved a significant longer PFS and OS versus non-responders (P = 0.007 and P = 0.003, respectively). The only grade 3-4 adverse event occurring in more than 5.0% of patients was neutropenia (6 patients, 7.5%). No statistically significant association was noted between polymorphisms of the genes analyzed and clinical outcome. CONCLUSIONS: In patients with platinum-pretreated, advanced NSCLC and favorable clinical prognostic factors, treatment with CBDCA/PMX is associated with a good clinical outcome and toxicity profile. None of the SNPs analyzed was found to be a useful predictor of treatment efficacy.