Machine learning integrates genomic signatures for subclassification beyond primary and secondary acute myeloid leukemia.

Although genomic alterations drive the pathogenesis of acute myeloid leukemia (AML), traditional classifications are largely based on morphology, and prototypic genetic founder lesions define only a small proportion of AML patients. The historical subdivision of primary/de novo AML and secondary AML has shown to variably correlate with genetic patterns. The combinatorial complexity and heterogeneity of AML genomic architecture may have thus far precluded genomic-based subclassification to identify distinct molecularly defined subtypes more reflective of shared pathogenesis. We integrated cytogenetic and gene sequencing data from a multicenter cohort of 6788 AML patients that were analyzed using standard and machine learning methods to generate a novel AML molecular subclassification with biologic correlates corresponding to underlying pathogenesis. Standard supervised analyses resulted in modest cross-validation accuracy when attempting to use molecular patterns to predict traditional pathomorphologic AML classifications. We performed unsupervised analysis by applying the Bayesian latent class method that identified 4 unique genomic clusters of distinct prognoses. Invariant genomic features driving each cluster were extracted and resulted in 97% cross-validation accuracy when used for genomic subclassification. Subclasses of AML defined by molecular signatures overlapped current pathomorphologic and clinically defined AML subtypes. We internally and externally validated our results and share an open-access molecular classification scheme for AML patients. Although the heterogeneity inherent in the genomic changes across nearly 7000 AML patients was too vast for traditional prediction methods, machine learning methods allowed for the definition of novel genomic AML subclasses, indicating that traditional pathomorphologic definitions may be less reflective of overlapping pathogenesis.

Risk and subtypes of secondary primary malignancies in diffuse large B-cell lymphoma survivors change over time based on stage at diagnosis.

BACKGROUND: Previous studies have shown an increased risk of secondary primary malignancies (SPMs) after diffuse large B-cell lymphoma (DLBCL) treatment. Whether stage of DLBCL at diagnosis affects the subtypes of SPMs that occur has not been previously described. METHODS: The Surveillance, Epidemiology, and End Results database was queried for patients aged >18 years diagnosed with primary DLBCL from 1973 to 2010 and categorized by early stage (ES) (stage I-II) or advanced stage (AS) (stage III-IV) disease. Differences in overall and location-specific SPM incidence by stage and time since diagnosis were assessed in 5-year intervals using a Fine-Gray hazards model. Overall survival was compared using the log-rank test. A Cox proportional hazards model was used to assess differences in survival. RESULTS: In total, 26,038 patients with DLBCL were identified, including 14,724 with ES and 11,314 with AS disease. The median follow-up was 13.3 years. Overall, 13.0% of patients developed SPM, with a higher but nonsignificantly increased risk of SPM development in those who had ES disease compared with those who had AS disease (14% vs 11.6%; P = .14). During the first 5 years after diagnosis, patients who had ES disease had a higher risk of SPM than those who had AS disease, specifically colorectal, pancreas, breast, and prostate SPMs. During the period from 10 to 15 years after diagnosis, patients who had AS disease had a higher risk of SPM than those who had ES disease, specifically hematologic SPMs. Development of SPM was found to significantly increase the risk of death regardless of stage at diagnosis. CONCLUSIONS: In this large, population-based study, distinctly different subtypes and temporal patterns of SPM development were identified based on stage of DLBCL at diagnosis. The current study merits consideration of tailored site-specific and time-specific surveillance for patients with DLBCL according to stage and time interval since diagnosis.

Patterns of Prior and Subsequent Neoplasms in Children and Adolescents With Soft Tissue Sarcomas.

BACKGROUND: The occurrence of prior, concurrent and subsequent neoplasms (SN) represents a serious problem in children and adolescents with soft tissue sarcomas. Pathogenic germline variants contribute to the diagnosis of multiple neoplasms in sarcoma survivors. MATERIALS AND METHODS: The records of 748 children and adolescents, diagnosed with soft tissue sarcomas and registered in trials/registries by the cooperative soft tissue sarcoma (Cooperative Weichteilsarkom Studie) group, were reviewed for the occurrence of SNs. Reference histology review was available for all cases; the presence of oncogenic fusions known at the time of diagnosis was confirmed for fusion-positive (F+) entities. RESULTS: Concurrent or subsequent SNs developed in 13 of 473 survivors of fusion-negative (F-) sarcomas, for an 8-year cumulative SN incidence of 5% in survivors of F- sarcomas. In contrast, only 1 of 278 survivors of F+ sarcoma developed an SN. Twenty of 748 patients with soft tissue sarcomas had a history of prior neoplasms. Six of 14 patients who developed SNs after their index sarcomas met Chompret criteria for Li-Fraumeni syndrome. Nine of 20 patients who had tumors before their index sarcoma diagnosis had neurofibromatosis type 1 or neurofibromatosis type 1 spectrum tumors. CONCLUSION: Sarcoma phenotype/genotype and the sequence and nature of prior and subsequent neoplasms provide a window into underlying germline genetic susceptibilities in children and adolescents with soft tissue sarcomas.

Impact of age at diagnosis of head and neck cancer on incidence of metachronous cancer.

BACKGROUND: Metachronous cancer in patients with head and neck cancer (HNC) is common and is associated with a poor prognosis. We aimed to evaluate the incidence of metachronous cancer at different sites according to age at diagnosis of index HNC. METHODS: We collected data on 2011 patients with oral cancer, oropharynx cancer, hypopharyngeal cancer, and laryngeal cancer as index cancers using the Osaka International Cancer Institute Cancer Registry database between 2005 and 2016. Among these, we analyzed 1953 patients after excluding 5 patients who were not followed-up and 53 patients with simultaneous multiple index cancers. We evaluated the cumulative incidence of metachronous cancer in the esophagus, lung, and other sites according to age at diagnosis of the index HNC using the Kaplan-Meier method. Multivariate logistic regression analysis was performed to identify factors that influenced the incidence of metachronous cancers following HNC. RESULTS: The cumulative incidence of metachronous esophageal cancer in young patients (< 65 years) was significantly higher than that in old patients (≥ 65 years) (12.1% vs 8.5% at 5 years, and 16.5% vs 11.2% at 10 years; p = 0.015). On the other hand, the cumulative incidence of the other cancers in young patients was significantly lower than that in old patients (7.8% vs 12.2% at 5 years, and 13.9% vs 15.3% at 10 years; p = 0.017). The cumulative incidence of lung cancer was not significance according to age at diagnosis of the index HNC. In the multivariate analysis, histological type (squamous cell carcinoma) and lesion location (hypopharynx and larynx) were independently associated with metachronous cancers. Moreover, age at diagnosis of the index HNC (< 65 years), histological type (squamous cell carcinoma) and lesion location (hypopharynx) were significant predictors of metachronous esophageal cancer incidence and lesion location (hypopharynx) was a significant predictor of metachronous lung cancer incidence. CONCLUSION: Risk stratification of metachronous cancers with age and other predictors may help to properly manage patients with HNC. TRIAL REGISTRATION: The present study is a non-intervention trial.

Second primary colorectal cancer after the initial primary colorectal cancer.

BACKGROUND: Initial primary colorectal cancer (IPCRC) has a high risk of developing into second primary colorectal cancer (SPCRC). Right-sided colon cancer (RCC) and left-sided colon cancer (LCC) have different characteristics and are considered to be two different entities. However, the different risks for SPCRC in categorized tumor sites and SPCRC subcategorized sites have not been fully elucidated to date. We aimed to compare incidence and survival of IPCRC and SPCRC and characterize the risk factors of SPCRC while also comparing the different SPCRC characteristics. METHODS: We used the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) data to compute standardized incidence ratios (SIR) in order to estimate risk of SPCRC after IPCRC diagnosis. The most prominent risk factors for SPCRC were measured by multivariate regression analysis and the temporal trend of SPCRC incidence was assessed with Joinpoint regression. Survival of patients with SPCRC and IPCRC was compared by Kaplan-Meier analysis. RESULTS: Patients with IPCRC were 1.73 times more likely to develop SPCRC (SIR = 1.73, 95% CI 1.69-1.78). SPCRC incidence declined since the first 8 years of IPCRC diagnosis to baseline. We demonstrated poorer survival with SPCRC compared with IPCRC while second RCC resulted in better survival compared with second LCC. Black ethnicity, age range 70-79, and LCC were associated with the highest risk of developing SPCRC. CONCLUSION: The characteristic differences between second LCC and RCC were relatively narrow. Furthermore, in those with SPCRC, RCC had the best survival outcome.

Type of second primary malignancy after achieving complete response by definitive chemoradiation therapy in patients with esophageal squamous cell carcinoma.

BACKGROUND: While the standard treatment for stage II-III (non-T4) esophageal squamous cell carcinoma (ESCC) is neoadjuvant therapy followed by esophagectomy, definitive chemoradiation therapy (dCRT) is an option to treat ESCC patients who reject or may not tolerate surgical treatment. Second primary malignancy (SPM) is a problem for long-term survivors after achieving complete response (CR) by dCRT. METHODS: The source of the subjects in this study was the patients with stage II/III (excluding T4 disease) ESCC (UICC6th) who underwent dCRT from 2000 to 2011 at the National Cancer Center Hospital, Japan. SPM, defined as malignancy newly detected at different site from the initial disease, was checked in patients who achieved CR by the initial dCRT. RESULTS: Among the 285 patients with stage II/III (excluding T4 disease) ESCC who underwent dCRT, 185 patients achieved CR. SPM was detected in 49 patients (median time to developing SPM, 41.5 months), accounting for 19.3% (95% CI 0.137-0.257) as the 5-year cumulative risk of SPM. SPMs were head and neck cancer (n = 12), gastric cancer (n = 12), esophageal cancer (n = 7), lung cancer (n = 5), colon cancer (n = 4), diffuse large B-cell lymphoma (n = 3), bladder cancer (n = 2), small intestinal cancer (n = 1), cholangiocarcinoma (n = 1), malignant melanoma (n = 1), and breast cancer (n = 1). There were no significant differences in baseline characteristics between the patients who developed SPM (n = 49) and others (n = 136). CONCLUSIONS: Because second primary malignancy developed often after achieving CR by dCRT for ESCC, it should be followed carefully.

Superior outcome of pediatric acute myeloid leukemia patients with orbital and CNS myeloid sarcoma: a report from the Children's Oncology Group.

BACKGROUND: Extramedullary leukemia (EML) is common in pediatric acute myeloid leukemia (AML) and occurs as leukemia cells within the cerebrospinal fluid (CSF) or as a solid tumor (myeloid sarcoma-MS). The effect of MS on survival is unknown. METHODS: Patients on CCG protocols 2861, 2891, 2941, and 2961 being treated for AML with intensive-timing chemotherapy were classified for the presence of EML (CSF leukemia, CNS-MS, orbital-MS, or non-CNS MS). CSF leukemia was classified as CNS3 (>5 WBC in the CSF with blasts) and non-CSF leukemia as CNS1/2 (<5 WBC in the CSF with or without blasts). Characteristics and outcomes of these patients were compared. RESULTS: Of the 1,459 total patients, 1,206 (82%) had no EML, 154 (11%) had CSF leukemia, 19 (1%) had CNS-MS, 23 (2%) had orbital-MS, and 57 (4%) had non-CNS MS. The CR rate was significantly higher in patients with orbital-MS and CNS-MS than in those with non-MS and non-CNS MS (96% and 95% vs. 78% and 78%, P = 0.034). Patients with orbital-MS and CNS-MS had significantly higher overall survival than patients with non-CNS MS (92% and 73% vs. 38%, P < 0.001), CNS3 patients (92% and 73% vs. 51, P < 0.001), and CNS1/2 patients (92% and 73% vs. 50%, P < 0.001). Patients with orbital-MS had a significantly lower relapse rate. CONCLUSION: Patients with MS involving orbital and CNS sites had a significantly better survival than patients with non-CNS MS, with CSF leukemia, or with no EML.

Identification of metastatic cancer in claims data.

PURPOSE: To develop algorithms to identify metastatic cancer in claims data, using tumor stage from an oncology electronic medical record (EMR) data warehouse as the gold standard. METHODS: Data from an outpatient oncology EMR database were linked to medical and pharmacy claims data. Patients diagnosed with breast, lung, colorectal, or prostate cancer with a stage recorded in the EMR between 2004 and 2010 and with medical claims available were eligible for the study. Separate algorithms were developed for each tumor type using variables from the claims, including diagnoses, procedures, drugs, and oncologist visits. Candidate variables were reviewed by two oncologists. For each tumor type, the selected variables were entered into a classification and regression tree model to determine the algorithm with the best combination of positive predictive value (PPV), sensitivity, and specificity. RESULTS: A total of 1385 breast cancer, 1036 lung, 727 colorectal, and 267 prostate cancer patients qualified for the analysis. The algorithms varied by tumor type but typically included International Classification of Diseases-Ninth Revision codes for secondary neoplasms and use of chemotherapy and other agents typically given for metastatic disease. The final models had PPV ranging from 0.75 to 0.86, specificity 0.75-0.97, and sensitivity 0.60-0.81. CONCLUSIONS: While most of these algorithms for metastatic cancer had good specificity and acceptable PPV, a tradeoff with sensitivity prevented any model from having good predictive ability on all measures. Results suggest that accurate ascertainment of metastatic status may require access to medical records or other confirmatory data sources.

Classification of ipsilateral breast tumor recurrences after breast conservation therapy can predict patient prognosis and facilitate treatment planning.

OBJECTIVE: To classify ipsilateral breast tumor recurrences (IBTR) as either new primary tumors (NP) or true local recurrence (TR). We utilized 2 different methods and compared sensitivities and specificities between them. Our goal was to determine whether distinguishing NP from TR had prognostic value. BACKGROUND: After breast-conservation therapy, IBTR may be classified into 2 distinct types (NP and TR). Studies have attempted to classify IBTR by using tumor location, histologic subtype, DNA flow cytometry data, or gene-expression profiling data. METHODS: A total of 447 (7.9%) of 5660 patients undergoing breast-conservation therapy from 1970 to 2005 experienced IBTR. Clinical data from 397 patients were available for review. We classified IBTRs as NP or TR on the basis of either tumor location and histologic subtype (method 1) or tumor location, histologic subtype, estrogen receptor status and human epidermal growth factor receptor 2 status (method 2). Kaplan-Meier curves and log-rank tests were used to evaluate overall and disease-specific survival differences between the 2 groups. Classification methods were validated by calculating sensitivity and specificity values using a Bayesian method. RESULTS: Of 397 patients, 196 (49.4%) were classified as NP by method 1 and 212 (53.4%) were classified as NP by method 2. The sensitivity and specificity values were 0.812 and 0.867 for method 1 and 0.870 and 0.800 for method 2, respectively. Regardless of method used, patients classified as NP developed contralateral breast carcinoma more often but had better 10-year overall and disease-specific survival rates than those classified as TR. Patients with TR were more likely to develop metastatic disease after IBTR. CONCLUSION: Ipsilateral breast tumor recurrences classified as TR and NP had clinically different features, suggesting that classifying IBTR may provide clinically significant data for the management of IBTR.

Patient and tumor characteristics may raise clinicians' awareness of familial colorectal cancer: a Norwegian population-based study.

BACKGROUND: The aim of this study was to identify patient and tumor characteristics that may be useful for clinicians in the identification of possible familial colorectal cancer (CRC). METHODS: Among 562 unselected, consecutively diagnosed colorectal cancer patients, 490 patients were included and divided into familial or sporadic CRC based on family history data collected by standardized patient interviews. Clinical data were collected from the patients' medical records. Patients were classified as familial CRC according to currently accepted family history criteria. Each patient was also classified according to Amsterdam II criteria and the 4th and 5th criteria in the Revised Bethesda Guidelines that address information on family history. Tumors were described by location and histopathology; 316 tumors were examined by molecular analyses. RESULTS: Right-sided colonic tumors and synchronous CRC tumors were associated with familial CRC, and patients with metachronous CRC or Lynch syndrome-related tumors were more likely to be classified as familial CRC. The patients' age at diagnosis did not differ between the groups. Other patient or tumor characteristics were not associated with familial CRC. CONCLUSION: CRC patients with right-sided colonic cancers, synchronous cancers and previous CRC and/or LS-related tumors were more likely to have familial CRC, and young age at diagnosis was not associated with familial CRC in these data. A detailed family history of colorectal cancer is essential in the identification of patients and families with familial CRC.

Secondary bone lesions in the affected limb in osteosarcoma (skip lesions), its classification and prognosis.

BACKGROUND: Secondary bone lesions in the affected limb (SLAL) could appear after the limb-saving operation for osteosarcoma. If limb-saving surgery with a reduced surgical margin is contributing to the appearance of SLAL and eventually make the prognosis worse, we should prefer more extensive surgery. On the other hand, if SLAL is closely related to the systemic metastases, SLAL should be treated mainly by chemotherapy and limb saving-operation is still preferable. METHODS: The authors analyzed their patients with SLAL in order to get the optimum guidelines for the surgical treatment. 116 patients with osteosarcoma arising in the extremities were retrospectively reviewed. Among them 106 primary lesions were resected with a margin of more than 5 cm apart from the lesion. RESULTS: Twelve patients (10%) showed SLAL in various timing of the patients' clinical course. All SLAL were accompanied with the distant metastases. In addition, a long survivor who had lung metastases showed SLAL appeared before lung metastasis. CONCLUSION: We concluded that, SLAL was equal to systemic metastases. Therefore, the surgical margin which is more than 5 cm apart from the primary lesion is not enough to prevent SLAL. Only intensive chemotherapy is suspected to be useful to treat the undetectable SLAL on initial examination.

Therapy-related myelodysplastic syndromes deserve specific diagnostic sub-classification and risk-stratification-an approach to classification of patients with t-MDS.

In the current World Health Organization (WHO)-classification, therapy-related myelodysplastic syndromes (t-MDS) are categorized together with therapy-related acute myeloid leukemia (AML) and t-myelodysplastic/myeloproliferative neoplasms into one subgroup independent of morphologic or prognostic features. Analyzing data of 2087 t-MDS patients from different international MDS groups to evaluate classification and prognostication tools we found that applying the WHO classification for p-MDS successfully predicts time to transformation and survival (both p < 0.001). The results regarding carefully reviewed cytogenetic data, classifications, and prognostic scores confirmed that t-MDS are similarly heterogeneous as p-MDS and therefore deserve the same careful differentiation regarding risk. As reference, these results were compared with 4593 primary MDS (p-MDS) patients represented in the International Working Group for Prognosis in MDS database (IWG-PM). Although a less favorable clinical outcome occurred in each t-MDS subset compared with p-MDS subgroups, FAB and WHO-classification, IPSS-R, and WPSS-R separated t-MDS patients into differing risk groups effectively, indicating that all established risk factors for p-MDS maintained relevance in t-MDS, with cytogenetic features having enhanced predictive power. These data strongly argue to classify t-MDS as a separate entity distinct from other WHO-classified t-myeloid neoplasms, which would enhance treatment decisions and facilitate the inclusion of t-MDS patients into clinical studies.

Evaluation of Pharyngeal Background Mucosa in Patients with Superficial Hypopharyngeal Carcinoma.

OBJECTIVES/HYPOTHESIS: Transoral surgery (TOS) has become increasingly popular for patients with superficial hypopharyngeal squamous cell carcinoma (SCC). However, the number of patients in whom metachronous multiple SCC of the head and neck (HNSCC) occurs has also increased. In this study, we investigated whether multiple lugol-voiding lesions (LVLs) in the pharyngeal background mucosa observed during TOS would be a biomarker of metachronous HNSCC. STUDY DESIGN: Retrospective study. METHODS: We examined 362 patients who underwent TOS for superficial hypopharyngeal carcinoma. Endoscopic images were reviewed in a blinded fashion by two endoscopists. LVLs in the pharyngeal mucosa were graded as follows: A, no lesions; B, 1 to 4 lesions; and C, ≥5 lesions per endoscopic view. RESULTS: Cumulative incidence curves of secondary HNSCC in the groups of grades A, B, and C revealed 3-year incidence rates of 14.4%, 18.8%, and 29.3%, respectively (P = .001 for A vs. C and P = .002 for B vs. C). Cumulative incidence curves of third HNSCC in the groups of grades A, B. and C revealed 5-year incidence rates of 3.9%, 9.8%, and 19.6%, respectively (P = .001 for A vs. C and P = .006 for B vs. C). Cumulative incidence curves of fourth HNSCC in the groups of grades A, B, and C revealed 7-year incidence rates of 0%, 2.3%, and 13.2%, respectively (P = .025 for A vs. C and P = .009 for B vs. C). CONCLUSIONS: Multiple LVLs in the pharyngeal mucosa increase the risk of development of metachronous multiple HNSCC. LEVEL OF EVIDENCE: 3 (nonrandomized, controlled cohort/follow-up study) Laryngoscope, 131:2036-2040, 2021.

Secondary malignant neoplasms after Wilms tumor: an international collaborative study.

A combined cohort of 8,884 North American, 2,893 British and 1,574 Nordic subjects with Wilms tumor (WT) diagnosed before 15 years of age during 1960-2004 was established to determine the risk of secondary malignant neoplasms (SMN). After 169,641 person-years (PY) of observation through 2005, 174 solid tumors (exclusive of basal cell carcinomas) and 28 leukemias were ascertained in 195 subjects. Median survival time after a solid SMN diagnosis 5 years or more from WT was 11 years; it was 10 months for all leukemia. Age-specific incidence of secondary solid tumors increased from approximately 1 case per 1,000 PY at age 15 to 5 cases per 1,000 PY at age 40. The cumulative incidence of solid tumors at age 40 for subjects who survived free of SMNs to age 15 was 6.7%. Leukemia risk, by contrast, was highest during the first 5 years after WT diagnosis. Standardized incidence ratios (SIRs) for solid tumors and leukemias were 5.1 and 5.0, respectively. Results for solid tumors for the 3 geographic areas were remarkably consistent; statistical tests for differences in incidence rates and SIRs were all negative. Age-specific incidence rates and SIRs for solid tumors were lower for patients whose WT was diagnosed after 1980, although the trends with decade of diagnosis were not statistically significant. Incidence rates and SIRs for leukemia were highest among those diagnosed after 1990 (p-trend = 0.003). These trends may reflect the decreasing use of radiation therapy and increasing intensity of chemotherapy in modern protocols for treatment of WT.

Synchronous and metachronous malignant tumours expect the unexpected.

OBJECTIVE: To evaluate occurrence of synchronous and metachronous malignant tumours, to find tumour types, age group, and relationship to treatment received. METHODS: Previously diagnosed 1st primary tumour cases experiencing a synchronous or metachronous tumour, seen at AOI from February 2003 to August 2009 (78 months) were included. The cases were analyzed for morphology/histology of 1st primary tumour, age and gender of patient, treatment received for first tumour, time interval between the 1st and 2nd primary tumour, morphology/histology of second tumour, and the treatment conferred for 2nd tumour. RESULTS: The 2nd synchronous and metachronous tumours were 46/4025 (1.14%), in 18 males and 28 females (M:F 1:1.6). The age range was 16-75 years (median 43 years). The follow up time was 24-150 months. The time to 2nd primary tumour was 2-132 months. The 1st primary tumours were breast, ovary, GIT and urinary bladder. The patients received surgery, radiotherapy, chemotherapy, and hormonal therapy alone or as multi-modality treatment for the 1st tumours. The frequent 2nd tumours were breast, ovary and Gastro Intestinal tumours. CONCLUSION: It is imperative that patients with a primary malignant tumour should be thoroughly, closely, and regularly followed. Genetic counseling, risk estimation, cancer screening and chemoprevention must be emphasized. Every subsequent occurring tumour should be biopsied. The effect of 1st tumour on the 2nd or vice versa are still not fully understood and need exploration. The 2nd primary tumour is usually more aggressive, treatment resistant, and metastasizes early requiring a more aggressive treatment strategy.

Assessment of Trends in Second Primary Cancers in Patients With Metastatic Melanoma From 2005 to 2016.

Importance: To date, the risk of developing second primary cancers (SPCs) after the first primary melanoma has not been studied in the era of immune checkpoint inhibitors (ICIs). Objective: To assess differences in the risk of SPCs in patients with primary melanoma before (2005-2010) and after (2011-2016) the introduction and approval of ICIs. Design, Setting, and Participants: Population-based cohort study using the Surveillance, Epidemiology, and End Results database from January 2005 to December 2016 of patients diagnosed with metastatic melanoma. Data were analyzed from January 4 to June 30, 2020. Exposures: Receipt of immunotherapy or other anticancer agents. Main Outcomes and Measures: The primary outcome was the development of second primary cancers in patients with melanoma. Standardized incidence ratios (SIRs) were calculated for the development of SPCs before and after the introduction of ICIs. Results: Among 5016 patients with diagnosed metastatic melanoma, 2888 (58%) were younger than 65 years at the time of diagnosis, and 3441 (69%) were male. From 2005 to 2010, SIRs were 3.24 (95% CI, 0.08-18.04) for small intestine cancer, 1.93 (95% CI, 1.14-3.05) for lung and bronchus cancer, 2.77 (95% CI, 1.02-6.03) for kidney cancer, and 7.29 (95% CI, 2.93-15.02) for myeloma. From 2011 to 2016, SIRs were 9.23 (95% CI, 1.12-33.35) for small intestine cancer, 1.54 (95% CI, 0.71-2.93) for lung and bronchus cancer, 2.66 (95% CI, 0.73-6.82) for kidney cancer, and 5.90 (95% CI, 1.61-15.10) for myeloma. The overall risk of developing SPCs in individuals who survived the first primary melanoma was 65% higher (SIR, 1.65; 95% CI, 1.35-2.00) in the pre-ICIs period and 98% higher (SIR, 1.98; 95% CI, 1.57-2.45) in the post-ICIs period than the overall cancer incidence rate in the general population. Conclusions and Relevance: In this study, an increase in the overall risk of second primary cancers after melanoma after the introduction of immune checkpoint inhibitors was observed. The pattern of SPCs has been altered in the era of systemic therapy. Close monitoring and screening for SPCs may be warranted in patients with metastatic melanoma.

Plasmablastic lymphoma: CNS involvement, coexistence of other malignancies, possible viral etiology, and dismal outcome.

The clinical and pathological findings of plasmablastic lymphoma (PBL) have been described in the literature but the etiology is not well established, and treatment options are poorly defined. We reviewed patients with PBL in our institution to characterize the clinicopathologic features in our patient population. In this retrospective analysis from a single academic institution, five patients with PBL were identified and analyzed. Human immunodeficiency virus and human herpesvirus 8 (HHV-8) were identified in 40% (two out of five) and 80% (four out of five) of these patients, respectively. Central nervous system (CNS) involvement was identified in four out of five (80%) patients. Interestingly, three out of five patients had a concurrent or preceding second primary malignancy including small lymphocytic lymphoma, endometrial cancer, and nonsmall cell lung cancer. Most of the patients had advanced disease and a poor performance status at diagnosis. Only two of the patients received systemic chemotherapy with an initial partial response. All five patients died; the median overall survival was 1 month. Our experience in patients with PBL indicates that CNS involvement is more common than reported in the literature. Coexistence of a second primary malignancy may be frequent, and prognosis remains dismal with standard lymphoma therapy. Lastly, the role of HHV-8 in the etiopathogenesis needs further trials.

Probabilistic classifiers and automated cancer registration: an exploratory application.

A test of the performance of two probabilistic classifiers (random forests and multinomial logit models) in automatically defining cancer cases has been carried out on 5608 subjects, registered by the Venetian Tumour Registry (RTV) during the years 1987-1996 and manually checked for possible second cancers that occurred during the 1997-1999 period. An eightfold cross-validation was performed to estimate the classification error; 63 predictive variables were entered into the model fitting. The random forest allows to automatically classify 45% of subjects with a classification error lower than 5%, while the corresponding error is 31% for the multilogit model. The performance of the former classifier is appealing, indicating a potential drop of manually checked cases from 1750 to 960 per incidence year with a moderate error rate. This result suggests to refine the approach and extend it to other categories of manually treated cases.