Indocyanine green fluorescence angiography of the reconstructed gastric tube during esophagectomy: efficacy of the 90-second rule.

By examining the reconstructed gastric tube during esophagectomy using indocyanine green fluorescence (ICG) angiography, we have established a '90-second rule' to confirm good blood perfusion at the anastomosis site. We examined the surgical outcome (rate of anastomotic leakage) of 70 consecutive patients who underwent esophagectomy with gastric tube reconstruction using ICG fluorescence angiography. All of the anastomoses were made in the area where less than 90 seconds was needed for enhancement using ICG fluorescence angiography (i.e. within the 90-second rule). In 18 cases for which the time until enhancement of the gastric tube tip exceeded 60 seconds, the anastomosis site was decided by reference to the ICG fluorescence angiogram, and the hypoperfused area was excised, and this significantly shortened the median time until enhancement of the gastric tube tip from 95.5 (60.0-204.0) seconds to 41.0 (9.0-77.0) seconds (P < 0.001). In three cases, the anastomosis was made at the site where more than 60 seconds was needed for ICG enhancement. In one case where ICG enhancement had taken 77 seconds, minor anastomotic leakage occurred. The overall rate of anastomotic leakage in this series was 1.4%. Blood flow in the reconstructed gastric tube is sufficient if the anastomosis is made in the area where ICG fluorescence angiography demonstrates enhancement within 60 seconds. Gastric tube necrosis can be avoided if the area showing an enhancement time exceeding 90 seconds is excised. The 90-second rule is a safe and effective method for deciding the site of anastomosis.

Should isolated peritoneal carcinomatosis from colorectal cancer be sub-classified into stage IVB in era of modern chemotherapy?

BACKGROUND: According to the 7th edition of the TNM staging system, stage IV metastatic colorectal cancer (CRC) at the time of initial diagnosis is sub-classified into stage IVA or IVB disease. Peritoneal carcinomatosis (PC), considered to have a dismal prognosis, is exclusively sub-classified into stage IVB, even though other metastases to a sole organ are sub-classified into stage IVA, which is considered to be associated with better survival. This retrospective study was undertaken to investigate the overall survival in metastatic CRC patients, focusing on PC patients. METHODS: We reviewed data on patients with metastatic CRC at initial diagnosis surgically treated between January 2006 and June 2011. A survival analysis was performed paying special attention to PC and sub-classifying patients with PC into three categories according to metastatic sites. RESULTS: There were 69 stage IVA patients (IVA group) and 83 stage IVB. Among stage IVB patients, 20 had isolated PC (PC-I group), 28 had PC with one or more other sites of metastasis (PC-II group), and 35 had at least 2 metastatic without peritoneal involvement (NPC group). Of 152 stage IV patients, 132 (87 %) underwent resection of the primary tumor and 19 (12 %) underwent radical resection of metastatic disease with microscopic free margins (R0 resection) including 5/20 (25 %) patients in the PC1 group. A total of 139 patients received oxaliplatin-based chemotherapy in a palliative (n = 125), neoadjuvant (n = 3), or adjuvant setting after R0 resection (n = 11). Compared with 36.6 months in the PC-I group, median survival was 32.5 months (P = 0.48) in the IVA group, 14.7 months (P = 0.07) in the PC-II group, and 12.9 months (P < 0.01) in the NPC group. CONCLUSIONS: The sub-classification of isolated PC into stage IVA instead of IVB might be more appropriate in the era of modern chemotherapy. Further investigation is warranted.

γδ T cells compose a developmentally regulated intrauterine population and protect against vaginal candidiasis.

This most comprehensive analysis to date of γδ T cells in the murine uterus reveals them to compose a unique local T-cell compartment. Consistent with earlier reports, most cells expressed a canonical Vγ6Vδ1 TCR, and produced interleukin (IL)-17A upon stimulation. Nonetheless, contrasting with earlier reports, uterine γδ T cells were not obviously intraepithelial, being more akin to sub-epithelial Vγ6Vδ1+ T cells at several other anatomical sites. By contrast to other tissues however, the uterine compartment also included non-Vγ6+, IFN-γ-producing cells; was strikingly enriched in young mice; expressed genes hitherto associated with the uterus, including the progesterone receptor; and did not require microbes for development and/or maintenance. This notwithstanding, γδ T-cell deficiency severely impaired resistance to reproductive tract infection by Candida albicans, associated with decreased responses of IL-17-dependent neutrophils. These findings emphasise tissue-specific complexities of different mucosal γδ cell compartments, and their evident importance in lymphoid stress-surveillance against barrier infection.

Mechanism of regulation of actin polymerization by Physarum profilin.

Physarum profilin reduces the rates of nucleation and elongation of F-actin and also reduces the extent of polymerization of actin at the steady state in a concentration-dependent fashion. The apparent critical concentration for polymerization of actin is increased by the addition of profilin. These results can be explained by the idea that Physarum profilin forms a 1:1 complex with G-actin and decreases the concentration of actin available for polymerization. The dissociation constant for binding of profilin to G-actin is estimated from the kinetics of polymerization of G-actin and elongation of F-actin nuclei and from the increase of apparent critical concentration in the presence of profilin. The dissociation constants for binding of Physarum profilin to Physarum and muscle actins under physiological ionic conditions are in the ranges of 1.4-3.7 microM and 11.3-28.5 microM, respectively. When profilin is added to an F-actin solution, profilin binds to G-actin which co-exists with F-actin, and then G-actin is dissociated from F-actin to compensate for the decrease of the concentration of free G-actin and to keep it constant at the critical concentration. At the steady state, free G-actin of the critical concentration is in equilibrium not only with F-actin but also with profilin-G-actin complex. The stoichiometry of 1:1 for the formation of complex between profilin and G-actin is directly shown by means of chemical cross-linking.

Control of actin filament length by phosphorylation of fragmin-actin complex.

Fragmin is a Ca2(+)-sensitive F-actin-severing protein purified from a slime mold, Physarum polycephalum (Hasegawa, T., S. Takahashi, H. Hayashi, and S. Hatano. 1980. Biochemistry. 19:2677-2683). It binds to G-actin to form a 1:1 fragmin/actin complex in the presence of micromolar free Ca2+. The complex nucleates actin polymerization and caps the barbed end of the short F-actin (Sugino, H., and S. Hatano. 1982. Cell Motil. 2:457-470). Subsequent removal of Ca2+, however, hardly dissociates the complex. This complex nucleates actin polymerization and caps the F-actin regardless of Ca2+ concentration. Here we report that this activity of fragmin-actin complex can be abolished by phosphorylation of actin of the complex. When crude extract from Physarum plasmodium was incubated with 5 mM ATP and 1 mM EGTA, the activities of the complex decreased to a great extent. The inactivation of the complex in the crude extract was not observed in the presence of Ca2+. In addition, the activities of the complex inactivated in the crude extract were restored under conditions suitable for phosphatase reactions. We purified factors that inactivated fragmin-actin complex from the crude extract. These factors phosphorylated actin of the complex, and the activities of the complex decreased with an increased level of phosphorylation of the complex. These factors, termed actin kinase, also inactivated the complex that capped the barbed end of short F-actin, leading to elongation of the short F-actin to long F-actin. Thus the length of F-actin can be controlled by phosphorylation of fragmin-actin complex by actin kinase.

Dynamic aspects of the contractile system in Physarum plasmodium. III. Cyclic contraction-relaxation of the plasmodial fragment in accordance with the generation-degeneration of cytoplasmic actomyosin fibrils.

Plasmodial fragments of Physarum polycephalum, excised from anterior regions of a thin-spread plasmodium, contracted-relaxed cyclicly with a period of 3-5 min. The area of the fragments decreased approximately 10% during contraction. In most cases, there was little endoplasmic streaming which indicates that contractions were synchronized throughout the fragment. By both polarized light and fluorescence microscopy, the organization and distribution of the cytoplasmic actomyosin fibrils in the fragments changed in synchrony with the contraction cycle. The fibrils formed during the contraction phase, and finally became a highly organized framework consisting of a three-dimensional network of numerous fibrils with many converging points (the nodes). During relaxation, the fibrils degenerated and disappeared almost completely, though some very weak fibrils remained near the nodes and the periphery. The results obtained by fluorometry of the fragments, stained with rhodamine-phalloidin, suggested that the G-F transformation of actin is not the main underlying process of the fibrillar formation.

The polymerization of actin: its role in the generation of the acrosomal process of certain echinoderm sperm.

When Asterias or Thyone sperm come in contact with egg jelly, a long process which in Thyone measures up to 90 microm in length is formed from the acrosomal region. This process can be generated in less than 30 s. Within this process is a bundle of microfilaments. Water extracts prepared from acetone powders of Asterias sperm contain a protein which binds rabbit skeletal muscle myosin forming a complex whose viscosity is reduced by ATP. Within this extract is a protein with the same molecular weight as muscle actin. It can be purified either by collecting the pellet produced after the addition of Mg(++) or by reextracting an acetone powder of actomyosin prepared by the addition of highly purified muscle myosin to the extract. The sperm actin can be polymerized and by electron microscopy the polymer is indistinguishable from muscle F-actin. The sperm actin was shown to be localized in the microfilaments in the acrosomal processes by: (a) heavy meromyosin binding in situ, (b) sodium dodecyl sulfate (SDS) gel electrophoresis of the isolated acrosomal processes and a comparison to gels of flagella which contain no band corresponding to the molecular weight of actin, and (c) SDS gel electrophoresis of the extract from isolated acrosomal caps. Since the precursor for the microfilaments in the unreacted sperm appears amorphous, we suspected that the force for the generation of the acrosomal process is brought about by the polymerization of the sperm actin. This supposition was confirmed, for when unreacted sperm were lysed with the detergent Triton X-100 and the state of the actin in the sperm extract was analyzed by centrifugation, we determined that at least 80% of the actin in the unreacted sperm was in the monomeric state.

Some properties of Physarum actinin. A regulatory protein of actin polymerization.

A factor termed Physarum actinin was isolated and partially purified from plasmodia of a myxomycete, Physarum polycephalum. When Physarum actinin was mixed with purified Physarum or rabbit striated muscle G-actin in a weight ratio of about 1 actinin to 9 actin and then the polymerization of G-actin induced, G-actin polymerized to the ordinary F-actin on addition of 0.1 M KCl. However, it polymerized to Mg-polymer on addition of 2 mM MgCl2. The reduced viscosity (etasp/C) of the Mg-polymer was 1.2 dl/g, about one-seventh of that of the F-actin (7.4 dl/g). The sedimentation coefficient of the Mg-polymer was 22.8 S, almost the same as that of the F-actin (29.4 S). The Mg-polymer showed the specific ATPase activity of the order of 1 . 10(-3) mumol ATP/mg actin per min. It was shown that Physarum actinin copolymerized with G-actin to form Mg-polymer on addition of 2 mM MgCl2. The molecular weights of Physarum actinin were about 90 000 in salt-free or slat solutions and 43 000 in a dodecyl sulfate solution. The range of salting out with ammonium sulfate was 50--65% saturation, which was different from that of Physarum actin (15--35% saturation). Physarum actinin did not interact with Physarum myosin or muscle heavy meromyosin. When the weight ratio of actinin to actin increased, the flow birefringence of the formed Mg-polymer decreased, and it became almost zero at the weight ratio of 1 actinin to 5 actin. ATPase activity reached the maximum level (2.2 . 10(-3) mumol ATP/mg actin per min) at the same ratio. On the addition of Physarum actinin to purified Physarum F-actin which had been polymerized on addition of 2 mM MgCl2 the viscosity decreased rapidly, suggesting that the F-actin filaments were broken in the smaller fragments or that they transformed to Mg-polymers. A factor with properties similar to Physarum actinin was isolated from acetone powder of sea urchin eggs.

Reversible superprecipitation and bundle formation of plasmodium actomyosin.

Synthetic actomyosin from plasmodium was found to undergo reversible superprecipitation upon addition of ATP. According to electronmicroscopic investigation upon clearing, short myosin filaments of about 0.2 micron in length appeared predominantly coexisting with actin filaments, and after superprecipitation, bundles of actin filaments were formed where short myosin filaments or myosin molecules were bound to the side of the bundle, making a whisk-like structure. The turbidity and the ATPase activity of actomyosin were measured at various ATP concentrations clamped by using an ATP-regenerating system. The turbidity was high below 1 . 10(-6) M ATP, corresponding to the state of superprecipitation, and with increasing ATP concentration it dropped in the range of 1 . 10(-6)--1 . 10(-5) M ATP. On the other hand, the ATPase activity was low below 1 . 10(-6) M ATP and increased above 1 . 10(-5) M after the turbidity dropped. Characteristic features of superprecipitation of plasmodium actomyosin observed here were discussed in relation to the mechanism of motility in vivo.

Mutational analysis of the PTEN/MMAC1 gene in non-Hodgkin's lymphoma.

The PTEN/MMAC1 gene at 10q23.3, which has dual specific phosphatase activity, is a novel tumor suppressor gene candidate. Various kinds of tumors have mutations in this gene, including glioblastoma, endometrial carcinoma and prostate cancer. We examined 29 cases of primary non-Hodgkin's lymphoma (NHL) for mutations in the PTEN/MMAC1 gene. One case of diffuse large B cell lymphoma had an 11 bp deletion, but the remaining 28 cases showed no mutations in the genome. Two of these 28 cases showed missense mutations in the PTEN/MMAC1 transcripts, but no alterations in the genomic DNA. These mRNA missense variants are similar to PTEN/MMAC1 transcript aberrations which have been reported in patients with breast cancer. These findings suggest that alterations in the PTEN/MMAC1 gene play a role in the pathogenesis of NHL.

Histone deacetylase inhibitors are the potent inducer/enhancer of differentiation in acute myeloid leukemia: a new approach to anti-leukemia therapy.

We investigated the effect of the histone deacetylase inhibitors (HDIs), trichostatin A and trapoxin A on leukemia cells and cell lines from the viewpoint of differentiation induction. TSA induced differentiation in erythroid cell lines by itself, whereas it synergistically enhanced the differentiation that was directed by all-trans retinoic acid (ATRA) or vitamin D3 in U937, HL60 and NB4 cells. The combined treatment of HDI with ATRA induced differentiation in ATRA-resistant HL60 and NB4 cells. The transcriptional expression during the treatment with HDI was examined in HL60, U937 and MEG-O1. Cell cycle-regulator genes (p21waf1 and p16INK4A) were upregulated or constantly expressed, erythroid-specific genes (GATA-1, beta-globin) were silent or downregulated, and housekeeping genes (beta-actin and GAPDH) were constantly expressed. Twelve of 35 (34%) clinical samples from AML patients ranging from M0 to M7 also displayed both phenotypical and morphological changes by the treatment with TSA alone. HDIs are thus the potent inducer or enhancer of differentiation in acute myeloid leukemia and regulate transcription in an ordered manner.

Identification of actin kinase activity in purified fragmin-actin complex.

Actin kinase phosphorylates actin of fragmin-actin complex, resulting in the inactivation of the nucleation and capping activities of the complex. Fragmin-actin complex was prepared by a new purification procedure. Incubation with ATP caused inactivation of the purified complex and phosphorylation of actin of fragmin-actin complex. The detailed analysis of the complex by SDS-gel electrophoresis showed that actin kinase was co-purified with the fragmin-actin complex. Formation of such an association between actin kinase and substrate suggests that the kinase is localized on the fragmin-actin complex to efficiently regulate actin cytoskeletons.

A novel plant nuclear gene encoding chloroplast ribosomal protein S9 has a transit peptide related to that of rice chloroplast ribosomal protein L12.

We have cloned a novel nuclear gene for a ribosomal protein of rice and Arabidopsis that is like the bacterial ribosomal protein S9. To determine the subcellular localization of the gene product, we fused the N-terminal region and green fluorescent protein and expressed it transiently in rice seedlings. Localized fluorescence was detectable only in chloroplasts, indicating that this nuclear gene encodes chloroplast ribosomal protein S9. The N-terminal region of rice ribosomal protein S9 was found to have a high sequence similarity to the transit peptide region of the rice chloroplast ribosomal protein L12, suggesting that these transit peptides have a common lineage.

Erythrocyte manganese concentration in healthy Japanese children, adults, and the elderly, and in cord blood.

This study showed differences in erythrocyte manganese concentrations in different age groups and in cord blood. All determinations were performed with a flameless atomic absorption spectrophotometer. The mean erythrocyte manganese concentration at one month of age was 435.1 +/- 118.7 ng/g Hb, three to four times higher than in adults. Thereafter, it decreased rapidly and was constant from 4 months to 11 yr of age. However, at 12 to 19 yr of age the concentration in males decreased (108.1 +/- 20.2 ng/g Hb) and was significantly lower than in females (141.7 +/- 25.5 ng/g Hb) (p less than 0.01). This sex difference became even greater in adults aged 20 to 40 yr, when it was 91.5 +/- 22.4 ng/g Hb in males and 141.2 +/- 19.8 ng/g Hb in females (p less than 0.001). The erythrocyte manganese concentration in adults was 31.6 +/- 9.2 ng/ml, somewhat higher than the value previously reported, which may reflect dietary habits in Japan or a racial difference.

Copper levels in plasma and erythrocytes in healthy Japanese children and adults.

This study showed differences in plasma and erythrocyte copper concentrations in children and adults. All determinations were performed with a flameless atomic absorption spectrophotometer. The mean plasma copper level at 1 month of age was significantly lower than in adults. After 1 month of age the mean concentration in plasma increased to its peak value at 2 to 5 yr of age, then decreased gradually with age. At 7 months to 10 yr of age, the copper levels in plasma were significantly higher than in adults. The erythrocyte copper levels at 1 month to 1 yr of age were significantly higher than in adults. The copper content of erythrocytes was highest at 2 to 6 months of age and then decreased gradually. The copper concentration of erythrocytes may reflect more accurately liver and total body copper levels than does the plasma copper level. There is less possibility of copper contamination in erythrocytes than in hair. Therefore, the measurement of erythrocyte copper concentration may well be a helpful index of the total copper status.

Zinc, copper, manganese, and selenium metabolism in thyroid disease.

This study was designed to evaluate trace metal metabolism in adults with thyroid diseases. Erythrocyte zinc values were significantly lower than normal in hyperthyroidism and higher in hypothyroidism. A significantly higher than normal urinary excretion of zinc was observed in hyperthyroidism. The mean concentrations of plasma and erythrocyte copper were significantly above normal in hyperthyroidism. Plasma selenium levels were significantly lower than normal in hyperthyroidism. No statistically significant difference was found in plasma zinc, erythrocyte manganese, or urine copper values between patients with thyroid diseases and healthy controls. The erythrocyte manganese content correlated well with thyroxine and triiodothyronine levels. Plasma prealbumin and retinol-binding protein correlated well with the erythrocyte zinc content but not with plasma zinc levels. There was no correlation between erythrocyte superoxide dismutase activity and erythrocyte copper or zinc concentrations. The results of this study suggest that the metabolism of zinc, copper, manganese, and selenium is abnormal in thyroid diseases.

Avoidance of bone marrow suppression using A-5021 as a nucleoside analog for retrovirus-mediated herpes simplex virus type I thymidine kinase gene therapy.

Gene therapy using the herpes simplex virus thymidine kinase (HSV-TK) gene combined with an anti-herpes drug, ganciclovir (GCV), has been applied for human diseases, especially for cancer treatment. However, bone marrow toxicity has been the most consistent adverse effect of GCV treatment in clinical settings. We evaluated the cytotoxic activity of a novel guanosine analog, (1'S,2'R)-9[[1',2'-bis(hydroxymethyl)cycloprop-1'-yl]methyl]guanin e (A-5021), against retrovirus-mediated HSV-TK gene-transduced human lung cancer cells. The bone marrow toxicity of A-5021 and GCV was studied by colony formation assay in both rodent and human bone marrow specimens. We demonstrated that A-5021 had potent cytotoxic activity equal to that of GCV against the retroviral vector-mediated HSV-TK gene-transduced lung cancer cell lines. Further, phosphorylated A-5021 could be transferred to neighboring cells, and this analog killed HSV-TK- neighboring cells, as was the case for GCV. In contrast, A-5021 did not exhibit an inhibitory effect on bone marrow progenitor cells and colony formation (the 50% inhibitory concentration of the colony-forming units in culture = >100 microg/mL in human bone marrow specimens and >66 microg/mL in rodent bone marrow specimens). These results indicate that A-5021 has potent cytotoxic activity as a nucleoside analog for gene therapy using HSV-TK gene, and can be used much more safely than GCV.

Chromatin reprogramming of male somatic cell-derived XIST and TSIX in ES hybrid cells.

In mammalian somatic cells, the X chromosome is active in XY males, whereas one X chromosome is inactivated in XX females. On the active male X chromosome, the XIST and TSIX genes are transcribed in undifferentiated cells of pre-implantation embryos (undifferentiated state) and then down-regulated upon cell differentiation (differentiated state). To explore the epigenetic mechanism involved in the on-off switching of XIST and TSIX transcription in the active X chromosome, male somatic cells were hybridized with male embryonic stem (ES) cells. Fluorescence in situ hybridization analysis revealed that the XIST gene derived from somatic cells was derepressed, as shown by the advent of two pinpoint signals. This was confirmed by strand-specific RT-PCR of XIST and TSIX genes. To analyze changes in chromatin structure in the promoter regions of XIST and TSIX derived from somatic cells, histone tail modifications were studied by chromatin immunoprecipitation analysis. Histones H3 and H4, which were hypoacetylated in the somatic cells, were hyperacetylated in the hybrid cells, and histone H3 lysine 4, which was hypomethylated in the somatic cells, was hypermethylated in the hybrid cells, indicating that the reactivation of XIST and TSIX was linked with chromatin modifications. In the telomeric region of DXPAS34, acetylation of histones H3 and H4 was dependent on reactivation of XIST and TSIX, whereas histone H3 lysine 4 was constitutively methylated independent of the transcriptional activity of those genes. We propose that the chromatin reprogramming is linked with the resetting of the memory found in the process of choosing an active X chromosome.

Changes in glucose transporter 2 and carbohydrate-metabolizing enzymes in the liver during cold preservation and warm ischemia.

In order to examine glucose metabolism in liver grafts during cold preservation (24 and 48 hr), warm ischemia (60 and 120 min), a combination of the two and reperfusion, the amount of protein and mRNA of glucose transporter 2 and the activities of enzymes in glycolysis (glucokinase, phosphofructokinase, pyruvatekinase), gluconeogenesis (glucose 6-phosphatase, fructose 1,6-bisphosphatase), and the pentose phosphate pathway (glucose 6-phosphate dehydrogenase) were measured. It appeared that glucose transport, the pentose phosphate pathway, and gluconeogenesis were maintained during cold preservation and warm ischemia. The activity of glucokinase significantly decreased from the control value of 1.33 +/- 0.23 IU/g protein to 0.70 +/- 0.17 (24 hr, P<0.05) and 0.57 +/- 0.12 (48 hr, P<0.01) only during cold preservation. However, the activity of phosphofructokinase significantly decreased from the control value of 4.37 +/- 0.06 IU/g protein to 2.67 +/- 0.15 (60 min, P<0.0001) and 1.53 +/- 0.06 (120 min, P<0.0001) only during warm ischemia. This indicates that glycolysis deteriorates during both cold preservation and warm ischemia and demonstrates further that the balance between glycolysis and gluconeogenesis shifts to gluconeogenesis. Even when cold preservation was combined with warm ischemia, the activity of glucokinase decreased only during cold preservation and the activity of phosphofructokinase decreased only during warm ischemia. Furthermore, these changes were time-dependent. It is suggested that they can be used as a clock to measure the durations of cold preservation and warm ischemia separately and that the magnitude of an ischemic injury to a liver and a liver graft's viability can be indirectly estimated before transplantation.

A case-control study of lung cancer with special reference to asbestos exposure.

A case-control study of lung cancer was carried out in Yokosuka City, Kanagawa Prefecture, the location of a pre-war Japanese Imperial naval factory and present site of a U.S. naval base. Cytologically or pathologically confirmed male fatal cases of lung cancer during the period of 1978 to 1982 in Yokosuka Kyosai Hospital were compared with a control group in the same hospital. Controls who died from causes other than cancer, pneumoconiosis, accident, or suicide were matched by age to the cases. Information that included occupational and smoking history was obtained by interviews with the families of the 96 cases and 86 controls. Major results were as follows: a) The relative risks of lung cancer associated with asbestos exposure and suspected exposure were 2.41 (p less than 0.05) and 1.56, respectively, after controlling for age and smoking history, and the relative risk associated with smoking was 6.01 (p less than 0.05) after adjusting for age and asbestos exposure. b) The age- and smoking-adjusted relative risks of lung cancer associated with asbestos exposure were 3.40 (p less than 0.01) and 1.72 for Kreyberg groups I and II, respectively. Significantly elevated relative risk associated with smoking history was demonstrated for Kreyberg group I, but not for group II, after controlling for age and asbestos exposure.