Survival trends of childhood cancer diagnosed during 1970-1994 in Piedmont, Italy: a report from the Childhood Cancer Registry.

BACKGROUND: The Childhood Cancer Registry of Piedmont (CCRP) started its activity in 1967. It is population based and covers the Piedmont Region (population 4,500,000; NW Italy). This article reports on time trends in survival after a childhood cancer diagnosed during 1970-1994. PROCEDURE: During 1970-1994, 2,329 incident cases were registered at CCRP on the basis of histological and/or clinical information, excluding 30 cases reported only by death certificate. Histological or hematological diagnosis was available for 2,067 cases. Vital status was assessed through the offices of the town of residence. At the end of follow-up, 1,202 cases were alive, 1,084 dead and 43 were not traceable. Survival was measured for the major diagnostic groups using both univariate and multivariate statistics. RESULTS: The 5-yr survival rate for acute lymphoblastic leukemia (ALL) improved regularly from 24.7% in 1970-1974 to 81.1% in 1990-1994, for acute nonlymphoblastic leukemia (ANLL) from 0% to 38.1%, for non-Hodgkin lymphoma (NHL) from 25.2% to 67.7%, for tumors of the central nervous system (CNS) (all types) from 33.4% to 75.9% and for Ewing tumor from 0% to 90%. Focusing on survival by period of diagnosis, the highest 5-year survival rate was observed for children diagnosed during 1985-1989 for medulloblastoma, neuroblastoma (NB), retinoblastoma, Wilms tumor, osteosarcoma, and rhabdomyosarcoma and for children diagnosed in 1990-1994 for the remaining sites. The trend over time was statistically significant for ALL, ANLL, NHL, CNS tumors, NB, and osteosarcoma as well as for all malignancies together. CONCLUSIONS: Population-based survival studies are useful complements to clinical studies. Survival results in the present study are similar to those presented for other European countries and the United States. For most types of neoplasm (except CNS) survival probability appears to stabilize 5-10 years after diagnosis.

Physical impairment and social life goals among adult long-term survivors of childhood cancer: a population-based study from the childhood cancer registry of Piedmont, Italy.

AIMS AND BACKGROUND: The study describes the health status and the attainment of life goals in the adult survivors of childhood cancer recorded at the Childhood Cancer Registry of Piedmont. METHODS AND STUDY DESIGN: A postal questionnaire was sent to the general practitioner of the 690 cases born before 1976 and alive in 1991 after at least 5 years from diagnosis. The answer was received for 485 (72.9%) included in the analyses. Items in the questionnaire were: sequelae related to cancer and its treatment, health-related quality of life (according to Bloom's criteria), educational level attained, and employment status. RESULTS: Vital and marital status were obtained for all 690 cases at the offices of the town of residence. No medical condition was reported for 309 cases (63.7%). The overall proportion with a high school or university education was compared to corresponding figures for Piedmont in 1991, adjusted by age, and was as high as in the general population. Similar results are observed for occupation. Patients of both genders were married less than expected. Patients with leukemia (112 cases), non-Hodgkin's lymphoma (34) or Hodgkin's lymphoma (52) were reported to have the highest quality of life. In contrast, patients with tumors of the central nervous system (151) had the highest frequency of sequelae and the lowest score for health-related quality of life. They-also presented the lowest educational achievement, the lowest proportion of employment and, among males, the lowest frequency of marriage. CONCLUSIONS: Our study shows a good social adjustment of adult survivors from childhood cancer, with the exception of central nervous system tumors. From the methodologic point of view, the present study shows the feasibility of surveillance surveys on health-related quality of life with the contribution of general practitioners.

Mortality of parents of children with cancer: a population-based study in Turin, Italy.

AIMS AND BACKGROUND: A cohort study was conducted in order to measure cause-specific mortality among parents of children recorded in the Childhood Cancer Registry of Piedmont. It is the first study carried out on the subject in southern Europe. METHODS AND STUDY DESIGN: The study comprised the 740 children resident in Turin and in whom a cancer had been diagnosed in the period 1967-1991. Nominal data were obtained for 723 fathers and 733 mothers. At the end of the follow-up in 1995, 645 fathers were alive, 68 dead and 10 untraceable. Corresponding figures for mothers were 700, 28 and 5. Cause of death was known for 91 of 96 parents. RESULTS: The period of observation of parents started on the birth of the index child, however mortality analyses were restricted to the period after 1965 because of limited availability of local reference rates. Total mortality was lower than expected among fathers (66 observed deaths vs 88.2 expected, P <0.05) and mothers (28 vs 31.4). Fathers showed deficits (not statistically significant) of lung neoplasms (4 vs 9.9), cardiovascular diseases (18 vs 27.1) and hepatic cirrhosis (2 vs 6.6). No statistically significant variations in mortality were observed with time from diagnosis or according to life status of the children. No cancer deaths occurred among the mothers of sarcoma cases whereas 1.9 were expected. CONCLUSIONS: The data do not indicate any increase in mortality from cancer or other causes end, on the contrary, show a reduction in mortality which was more evident for the causes related to life style.

[Pediatric Tumor Registry of Piedmont. Descriptive epidemiology of malignant tumors in children in Piedmont, 1976-1994]. / Il Registro dei Tumori Infantili del Piemonte. Epidemiologia descrittiva dei tumori maligni in età pediatrica in Piemonte, 1967-1994.

Since 1967, the Childhood Cancer Registry of Piedmont measures cancer incidence and lethality among children aged 0-14 residents in the Region. Two thousand seven hundred twenty cases were recorded in the period 1967-94. Males were 55.4%. The highest frequency was observed in the age class 0-4 including 41.3% of cases. The most frequent malignancies were: Acute Lymphocytic Leukemias, CNS Tumours and Lymphomas. Incidence rates showed limited variation, both for total neoplasm and for the largest diagnostic categories. On the contrary, lethality decreased markedly: rate (per million children years) was 77.2 in 1967-69 and 59.4 in 1988-94. This trend as more evident for acute leukemias and CNS tumours. Better diagnostic techniques, anticancer and support therapies are the likely explanation for the improved prognosis. Prevalence increased, as a consequence of improved survival and curability: in the Province of Torino it increased from 62 cases per 100,000 children (age 0-14) in 1980 to 98 cases in 1994.

Incidence of second primary malignancies after a malignant tumor in childhood: a population-based survey in Piedmont (Italy).

We have studied the frequency of second primary malignancies (SPM) among the 2,328 children registered in 1967-1969 at the Childhood Cancer Registry of Piedmont, the largest population-based childhood cancer registry in Southern Europe. Since the population of Piedmont is not served by a conventional cancer registry covering all ages, SPMs were identified through a number of ad hoc surveys within a variety of sources. Eighteen SPM (all histologically diagnosed) were observed after a thorough survey conducted in the ontological departments in Piedmont and after a postal questionnaire addressed to general practitioners. Death certificates were also examined. The crude incidence rate was 116.5 per 100,000 person-years. Risk was higher among children whose first malignancy was diagnosed more recently (SIR = 9.8 for diagnoses in 1983-1989 vs. 4.5 for diagnoses in 1967-1974). The same tendency was confirmed in analyses restricted to children in whom leukemia was diagnosed as the first cancer. Clinical data regarding the treatment of the first malignancy were available for 16 children out of 18: 15 had received chemotherapy and 12 radiotherapy (9 SPM originated in the irradiation field). The interest of measuring the risk of SPM on a population basis (and not only in clinical series) and the advantage of close cooperation between epidemiologists and clinical oncologists are underlined.

Time trends in survival of children with acute lymphocytic leukemia in Piedmont, Italy: a report from the Population-Based Cancer Registry.

AIMS AND BACKGROUND: The Childhood Cancer Registry of Piedmont (RTI) periodically updates the life status of each registered child. Given its size, the RTI is the major (albeit geographically limited) Italian source of population-based survival rates of cancer in children. The present report describes time trends in survival of children with acute lymphocytic leukemia (ALL). METHODS: During 1970-87, 429 residents in Piedmont aged 0-14 were diagnosed as having ALL: they have been followed up until 1991. RESULTS: Five-year survival rates increased from 21% to 72% for children diagnosed ALL respectively in 1970-72 and 1985-87. Major improvements occurred up to the mid-seventies and again between cases diagnosed in the early and late eighties. Improvement in survival was statistically significant for children belonging to classes comprised between 2 and 10 years of age at diagnosis. Period of diagnosis was unrelated to probability of survival among the 13 cases diagnosed ALL at age 0. Survival was unrelated to sex, even in the early seventies and even after consideration of children dying more than 5 years after diagnosis. Between 1976-81 and 1982-87, an improvement in survival was found in all categories of WBC count at diagnosis: the ratio between the two estimates was somewhat higher for children with more than 50,000 WBC/mm3 at diagnosis than for other children. CONCLUSIONS: Present data are compared with those resulting from other population-based series: this exercise can be useful for an overall evaluation of delivery of cancer therapy at the population basis.

Childhood cancer registry of the Province of Torino, Italy. Survival, incidence, and mortality over 20 years.

Incident childhood cancers in the Province of Torino, Italy (population aged 0 to 14 years averaging 0.5 million) have been registered since 1967. Four population-based exhaustive surveys have been done to collect cases diagnosed during the periods 1967 to 1969, 1970 to 1975, 1976 to 1981, and 1982 to 1986. For each registered child, vital status on June 30, 1988 was assessed. This article reports incidence, mortality, and survival rates over a 20-year period. A statistically significant trend toward an increased incidence of soft tissue sarcomas was identified. Statistically insignificant trends included an increased incidence of brain tumors (probably reflecting improved diagnostic procedures) and a decreased incidence of thyroid tumors. Incidence rates of leukemias in the first year of life tended to decrease. As expected, survival rates of some childhood cancers dramatically improved throughout the 20-year period; this occurred in leukemias, brain tumors, soft tissue sarcomas, and renal tumors. Survival rates are compared with observations in comparable population-based series.

Cancer mortality among relatives of children with soft-tissue sarcoma: a national survey in Italy.

Information was obtained on the living status or cause of death of 2223 close relatives of 195 children with soft-tissue sarcomas (STS) diagnosed under age 15. Three-hundred nine relatives had died, from all causes, before STS diagnosis in the index child. The expected figure estimated from age- and sex-specific mortality rates in Italy was 293.3. Cancer was reported as cause of death in 76 relatives (75.1 expected). Seven grandmothers, 2 aunts, 1 uncle and 0 mothers died from breast cancer vs. 4.6, 0.9, 0.0 and 0.2 expected. Three siblings died from cancer (0.2 expected, P less than 0.01), i.e. STS, ependymoma and non-Hodgkin lymphoma. These results confirm and expand previous observations that STS in children are associated with other cancers, particularly childhood and breast cancer, in members of the same family.

Neurofibromatosis and malignant childhood cancers: a survey in Italy, 1970-83.

Neural tumors, Wilms' tumor, rhabdomyosarcoma and several types of leukemia have been previously described in association with neurofibromatosis (NF). In a nation-wide collection of cases in Italy, 15 children (0-14 years of age) with NF and cancer or leukemia were identified; 13 of them had been diagnosed with cancer between 1976-83. The expected number of children with cancer and NF in 1976-83 was 4.48. The distribution of tumor types was different from that found in the general population, with a higher proportion of tumors of neural crest origin as well as soft tissue sarcomas. In 7/15 the family history was positive for NF; in 5/7 the individuals affected included the mother and/or a maternal relative.

Results of a multicenter retrospective study on pediatric brain tumors in Italy.

This retrospective study was undertaken to evaluate the clinical characteristics, course and treatment of children (0-14 years of age) diagnosed with a primary CNS tumor during the period 1976-1982 in Italy. Four hundred and sixty-two patients (263 males and 199 females) were followed by 18 various neurosurgical and pediatric oncology centers. The histologic types most frequently reported were: medulloblastoma (23%), astrocytoma (16%), ependymoma (11%) and spongioblastoma (11%). Of the 388 patients who underwent surgery, radical excision was reported in 42%, partial excision in 32%, biopsy only in 6%, and unqualified surgery in 4%; 19% had no surgery. Radiotherapy and chemotherapy combined were administered in 61% of the 143 patients followed at pediatric oncology centers; 19% received radiotherapy alone, 3% chemotherapy alone, and 17% neither treatment. Forty-six percent of the patients were reported alive, 40% dead, and 14% lost to follow-up. Performance status was identified for 62 patients. The investigation revealed marked differences in the therapeutic treatment administered, thus precluding valid data analysis. This emphasizes the need to coordinate efforts among the institutions and the disciplines involved in the treatment of this form of childhood cancer.

Childhood Cancer Registry of the Province of Torino: survival patterns since 1967 and update of incidence rates.

During 1967-1981, the population-based Childhood (ages 0-14 years) Cancer Registry of the Province of Torino has recorded 1057 resident children with incident cancer. The life status of each child has been ascertained as of December 31, 1983. Sex- and age-specific incidence rates have been stable throughout the period. Annual mortality rates (per million) from leukemias (all types) decreased from 34.0 in 1967-69 to 19.4 in 1979-81. Correspondingly, the survival percentage cumulative of leukemic children at 3 years after diagnosis increased from 16 for children diagnosed in 1967-69 to 57 for those diagnosed in 1979-81. For other cancer types, no increases or debatable decreases in mortality rates and increases in survival rates were recognized.