A unified framework for constructing, tuning and assessing photometric redshift density estimates in a selection bias setting.

Photometric redshift estimation is an indispensable tool of precision cosmology. One problem that plagues the use of this tool in the era of large-scale sky surveys is that the bright galaxies that are selected for spectroscopic observation do not have properties that match those of (far more numerous) dimmer galaxies; thus, ill-designed empirical methods that produce accurate and precise redshift estimates for the former generally will not produce good estimates for the latter. In this paper, we provide a principled framework for generating conditional density estimates (i.e. photometric redshift PDFs) that takes into account selection bias and the covariate shift that this bias induces. We base our approach on the assumption that the probability that astronomers label a galaxy (i.e. determine its spectroscopic redshift) depends only on its measured (photometric and perhaps other) properties [Formula: see text] and not on its true redshift. With this assumption, we can explicitly write down risk functions that allow us to both tune and compare methods for estimating importance weights (i.e. the ratio of densities of unlabelled and labelled galaxies for different values of [Formula: see text]) and conditional densities. We also provide a method for combining multiple conditional density estimates for the same galaxy into a single estimate with better properties. We apply our risk functions to an analysis of ≈106 galaxies, mostly observed by Sloan Digital Sky Survey, and demonstrate through multiple diagnostic tests that our method achieves good conditional density estimates for the unlabelled galaxies.

Routine pulmonary function tests during bleomycin therapy. Tests may be ineffective and potentially misleading.

Forced vital capacity (FVC) and diffusing capacity for carbon monoxide (DLco) studies in two patients with proved interstitial fibrosis after low doses of bleomycin were contrasted with these measurements in 21 patients who received bleomycin without toxicity. The FVC decreased significantly (20% or more) in both fibrosis patients and in seven others. In one fibrosis patient and in three others the results could be explained by weakness. The DLco fell in both patients with fibrosis and in 12 others. Correction for decrease in hemoglobin level accounted for the change in one of the toxic patients and seven of the others, but hemoglobin correction made two insignificant changes significant and increased therapy values above control four times. Since these tests have many false-positive results and can be affected by weakness and anemia leading to false-negative results their use is ineffective and may be misleading.

Phase II study of mitomycin-C, vincristine, and bleomycin in advanced squamous cell carcinoma of the uterine cervix.

Utilizing the stathmokinetic principle of timed vincristine and bleomycin, we combined these two agents with Mitomycin-C. The dose schedule included vincristine 0.5 mg/m2 intravenously (i.v.) geginning on day 1 and repeated twice weekly for 12 weeks; each injection was followed in 6-12 hours by bleomycin 6 mg/m2 for 12 weeks. Mitomycin-C was administered as a 20 mg/m2 bolus beginning on day 2 and repeated at 6-week intervals. Thirty patients were entered into this study, 27 were fully available for response. Thirteen patients (48%) met criteria of response (greater than 50% reduction in volume of measurable tumor). Significant myelosuppression resulted from this therapy. Median leukopenia nadir was 3.8 X 10(3) cells/mm3 and median thrombocytopenia nadir was 116 X 10(3) cells/mm3. Additional toxic reactions included anemia, lassitude, anorexia, peripheral neuropath fever, and skin rash. Despite significant, but manageable, toxicity, this combination appears to represent an improvement in the chemotherapy of a traditionaly refractory solid tumor.

Combination of adriamycin and cyclophosphamide in the treatment of metastatic prostatic carcinoma: a phase II study.

A prospective trial of Adriamycin and cyclophosphamide was undertaken in the treatment of patients with hormonally refractory prostatic carcinoma. Of the 20 patients evaluable, three (15%) had a partial remission and five (25%) had stable disease for a total response of 40%. All patients who responded had subjective improvement consisting of weight gain (greater than or equal to 10% of body weight), improvement of performance status (greater than or equal 10% in Karnofsky index), and a decreased need for pain medication.

Phase II study of profiromycin vs mitomycin-C utilizing acute intermittent schedules.

A randomized prosective study of Mitomycin-C and its N-methyl derivative, Porfiromycin, was conducted. Thirty-two patients with disseminated gastrointestinal cancer or other disseminated abdominal adenocarcinoma were treated with Mitomycin-C; 31 patients received Porfiromycin. Both drugs were given by acute intermittent bolus schedule (Mitomucin-C , 22.5 mg/M2 or Porfiromycin, 75 mg/M2 every 6--8 weeks as a single bolus i.v. injection). Eleven patients (34%) who received Mitomycin-C entered into partial remission. In 10 of the 31 patients (32%) receiving Porfiromycin, partial remission occured. Analysis by tumor type demonstrated that in the Mitomycin-C treated group responses occured in 4 of 12 patients with colorectal carcinoma, in 4 of 9 with upper GI cancers, and in 3 of 11 with ovarian cancer. Correspondingly in Porfiromycin group responses occured in 2 of 12 colorectal carcinoma patients, in 3 of 7 upper GI cancer patients, and in 5 of 12 ovarian cancer patients. Both drugs produced significant myelosuppression; however, Porfiromycin toxicity appeared more cumulative. Further clinical trial of Mitomycin in an acute intermittent bolus schedule appears justified.

Combination chemotherapy with cis-diamminedichloroplatinum, oncovin, and bleomycin (COB) in advanced head and neck cancer: phase II.

Twenty-eight consecutive patients with advanced head and neck epidermoid carcinoma were treated with a single course of combination chemotherapy which included high dose cis-platinum II, vincristine, and bleomycin. All patients except one were evaluable for drug response and toxicity. The overall toxicity was mild and tolerable except in one patient who had fatal nephrotoxicity. Thirteen patients (48%) had a partial regression, and five with minimal response did improve, giving an overall response rate of 67%. More responses were noted in previously untreated patients, especially those with oral or hypopharyngeal tumors without distant metastases. Definite histological changes were seen in a few patients in remission. Objective tumor response was associated with subjective improvement, and possibly, a prolonged survival.

Pleural effusion in cancer patients. A prospective randomized study of pleural drainage with the addition of radioactive phsophorous to the pleural space vs. pleural drainage alone.

Sixty-seven patients with disseminated cancer were randomly allocated to treatment with continuous closed chest drainage removing all fluid for 72 hours (PD) or pleural drainage for 72 hours with the instillation into the pleural space of radioactive colloidal chromic phosphate (PD + 32P). Forty-nine patients had breast carcinoma, and the remaining 18 patients had other cancers. Four of 49 patients with breast cancer and 13 of 18 with other cancer were dead in 8 weeks from the onset of effusion. In the group of patients with breast cancer PD + 32P controlled the effusion in 12 of 22 (54%) and PD alone in 15 of 30 episodes (50%). In the nonbreast group of patients PD + 32P controlled the effusion in five of six evaluable episodes (83%), and PD alone was successful in two of nine (22%). In 33% of breast cancer patients and 25% of the nonbreast-cancer patients, systemic chemotherapy produced objective remissions. Pleural effusion did not recur in any of these patients.

Topical instillation of doxorubicin hydrochloride in the treatment of recurring superficial transitional cell carcinoma of the bladder.

Thirteen patients with recurring stage A transitional cell carcinoma of the bladder were treated with monthly bladder instillations of doxorubicin hydrochloride. Follow-up at 3-month intervals has shown excellent control of tumor recurrence with complete remissions for up to 21 months in 2 patients. The median for remission in 8 of 13 patients has been 10 months. Two patients with bladder papillomatosis were treated successfully by the same method.

The effects of topical instillation of adriamycin in bladder tumors of rats fed with FANFT.

Topical bladder instillation of adriamycin was evaluated in FANET produced rat tumors produced by diets containing (N-[4-(5-Nitro-2-Furyl)-2-Thiazolyl] Formamide). The drug was ineffective in either preventing or eradicating tumors. The failure of response in this animal model may be related to drug schedule, biological potential of this tumor, or ineffectiveness of Adriamycin in this tumor.

Phase 1-11 study of DTIC and cyclocytidine in disseminated malignant melanoma.

Seventeen patients with disseminated malignant melanoma were treated with DTIC (150 mg/m2, Days 1-5) and cyclocytidine (increasing doses sc, Days 1-10) in a phase I-II study. There was one early death. The remaining 16 patients were evaluable for response and toxicity. Two patients (13%) had CR lasting 8+ and 2+ months, while one patient (6%) had a PR lasting 1 month. Nausea and vomiting was seen in seven patients (44%), jaw pain in four (25%), and orthostatic hypotension in two (13%). Hematologic toxicity was not excessive, nor was it cumulative. The overall response rate of 19% was comparable to that reported with DTIC alone. This drug combination does not appear to offer any therapeutic advantage within the dosage range tested in disseminated malignant melanoma.

Phase I clinical trial of combined therapy with vinblastine (NSC-49842), bleomycin (NSC-125066), and cisdichlorodiammineplatinum(II)( NSC-119875).

Seventeen patients with various histologic types of incurable malignant disease were treated with a combination of vinblastine, bleomycin, and cis-dichlorodiammineplatinum(II). Creatinine and blood urea nitrogen elevations were noted but were not of a severe degree. White blood cell and platelet count depressions were seen and appeared to be cumulative, though not life-threatening. Tinnitus and high-frequency hearing loss were noted. Tumor regression was seen in one patient with adenocarcinoma of the lung and in one patient with a testicular tumor. This appears to be a manageable drug combination with frequent monitoring of renal, hematopoietic, pulmonary, and auditory function. A phase II study establishing the therapeutic efficacy of this combination in advanced testicular neoplasms now appears to be indicated.

Combination chemotherapy with cyclophosphamide, vincristine, methyl-CCNU, and bleomycin in advanced bronchogenic carcinoma: experience with 106 patients.

A phase II study was conducted with three different combinations of cyclophosphamide, Oncovin (vincristine), methyl-CCNU, and bleomycin (COMB) in 106 patients with advanced bronchogenic carcinoma of all histologies. The response rates were 27.3% for COMB I (six of 22 patients), 18.2% for COMB II (eight of 44 patients), and 7.5% for vincristine, methyl-CCNU, and bleomycin (three of 40 patients). Complete responses were seen only in the small cell carcinoma group. Duration of remission and overall survival were of significance for the small cell group, whereas survival for the other histologic types comparing responders to nonresponders was of no statistical significance. Toxicity was significant and included nausea and vomiting, neuropathy, and myelosuppression. These drug combinations were disappointing and only in small cell carcinoma was meaningful antitumor activity seen.

The practicality of chronic hepatic artery infusion therapy of primary and metastatic hepatic malignancies: ten-year results of 124 patients in a prospective protocol.

Ten-year results are presented of 124 patients with malignancy apparently limited to the distribution of the hepatic artery, treated to prospective protocol with continuous infusion of 5-FUdR through an hepatic artery catheter. Nearly all patients had moderate to massive hepatic replacement. Of 88 patients with colorectal carcinoma, 64 (73%) had clinically objective and subjective remission. Median survival for responders was 13 months; for the entire group, ten months. Of 13 patients with hepatoma, nine had clinically significant regression with a median survival of 11 months. Ten patients had carcinoma of the gall bladder or bile duct with seven obtaining clinically significant regression. Complications encountered are discussed and are similar to other series. Of the patients experiencing clinically significant remission, all but one reached the complete independence performance status, and 84% reached normal activity levels. Thus, for hepatic localized tumor, this therapy is worthwhile and practical.