[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Image description not available.
Figure 1.— Representative examples of loss of heterozygosity (LOH) and microsatellite alteration (MA) in human immunodeficiency virus–associated lung carcinomas. Each panel demonstrates microsatellite marker analysis of microdissected DNA of paired nonmalignant (N) and tumor (T) tissue samples from individual patients. A, Tetranucleotide repeat marker D3S1537 (3p22-24.2) shows LOH of the upper allele in the tumor sample. B, Tetranucleotide repeat marker D3S1351 (3p22-24.2) shows loss of 1 tumor allele and a mobility shift of the remaining allele. C, The TP53 pentanucleotide repeat marker shows a mobility shift of the lower tumor allele. D, Tetranucleotide repeat marker D3S2432 (3p22-24.2) shows a mobility shift of the upper tumor allele. E, Tetranucleotide repeat marker D3S1537 (3p22-24.2) shows a single allele in the nonmalignant tissue and an additional tumor allele ("gain of heterozygosity"). F, Dinucleotide repeat marker D3S1478 (3p21) shows a single allele in the nonmalignant tissue that has undergone a mobility shift in the tumor sample.
Image description not available.
Figure 2—Comparison of loss of heterozygosity (LOH) and microsatellite alteration (MA) in human immunodeficiency virus (HIV)-associated and sporadic lung carcinoma. A, Fractional regional loss index (an indicator of LOH at all chromosomal regions analyzed per case). B, Microsatellite alteration (MA) index (an indicator of the total frequency of MAs per case).
Image description not available.
Figure 3.—Frequency of microsatellite alteration (MA) (in all tumors) of individual microsatellite markers (identified on the x-axis), expressed as a fraction. Of the polynucleotide repeat markers, 4 (D3S2432, D3S1351, D3S1537, and ITIH-1) are tetranucleotide repeats and 1 (TP53) is a pentanucleotide repeat.
Table 1.—Clinicopathological and Molecular Data of HIV-Associated Male Lung Cancer Cases*
Image description not available.
Table 2.—Loss of Heterozygosity (LOH) at All Chromosomal Regions Analyzed
Image description not available.
Table 3.—Nonmalignant Tissue Analyzed for MA in 2 Cases With HIV-Associated Lung Carcinoma*
Image description not available.
1.
Biggar RJ, Rabkin CS. The epidemiology of AIDS-related neoplasms.  Hematol Oncol Clin North Am.1996;10:997-1010.
2.
Remick SC. Non–AIDS-defining cancers.  Hematol Oncol Clin North Am.1996;10:1203-1213.
3.
Knowles DM. Etiology and pathogenesis of AIDS-related non-Hodgkin's lymphoma.  Hematol Oncol Clin North Am.1996;10:1081-1111.
4.
Miles SA. Pathogenesis of AIDS-related Kaposi's sarcoma.  Hematol Oncol Clin North Am.1996;10:1011-1021.
5.
Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1997.  CA Cancer J Clin.1997;47:5-27.
6.
Parkin DM, Pisani P, Lopez AD, Masuyer E.  At least one in seven cases of cancer is caused by smoking: global estimates for 1985.   Int J Cancer.1994;59:494-504.
7.
Fraire AE, Awe RJ. Lung cancer in association with human immunodeficiency virus infection.  Cancer.1992;70:432-436.
8.
Sridhar KS, Flores MR, Raub WJ, Saldana M.  Lung cancer in patients with human immunodeficiency virus infection compared with historic control subjects.   Chest.1992;102:1704-1708.
9.
Karp J, Profeta G, Marantz PR, Karpel JP. Lung cancer in patients with immunodeficiency syndrome.  Chest.1993;103:410-413.
10.
Vaccher E, Tirelli U, Spina M.  et al.   Lung cancer in 19 patients with HIV infection: the Italian Cooperative Study Group on AIDS and Tumors (GICAT).   Ann Oncol.1993;4:85-86.
11.
Fishman JE, Schwartz DS, Sais GJ, Flores MR, Sridhar KS.  Bronchogenic carcinoma in HIV-positive patients: findings on chest radiographs and CT scans.   AJR Am J Roentgenol.1995;164:57-61.
12.
White CS, Haramati LB, Elder KH, Karp J, Belani CP.  Carcinoma of the lung in HIV-positive patients: findings on chest radiographs and CT scans.   AJR Am J Roentgenol.1995;164:593-597.
13.
Parker MS, Leveno DM, Campbell TJ, Worrell JA, Carozza SE. AIDS related bronchogenic carcinoma: fact or fiction?  Chest.1998;113:154-161.
14.
Irwin LE, Begandy MK, Moore TM.  Adenosquamous carcinoma of the lung and the acquired immunodeficiency syndrome.   Ann Intern Med.1984;100:158.
15.
Knudson AG. Hereditary cancers: clues to mechanisms of carcinogenesis.  Br J Cancer.1989;59:661-666.
16.
Gazdar AF, Bader S, Hung J.  et al.   Molecular genetic changes found in human lung cancer and its precursor lesions.   Cold Spring Harb Symp Quant Biol.1994;59:565-572.
17.
Minna JD, Sekido Y, Fong K, Gazdar AF. Molecular biology of lung cancer.  In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer: Principles and Practice of Oncology. 5th ed. Philadelphia, Pa: JB Lippincott; 1997:849-857.
18.
Weber JL, May PE.  Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction.   Am J Hum Genet.1989;44:388-396.
19.
Shibata D. Molecular tumor clocks and dynamic phenotype.  Am J Pathol.1997;151:643-646.
20.
Strand M, Prolla TA, Liskay RM, Petes TD.  Destabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair.   Nature.1993;365:274-276.
21.
Liu B, Parsons R, Papadopoulos N.  et al.   Analysis of mismatch repair genes in hereditary non-polyposis colorectal cancer patients.   Nat Med.1996;2:169-174.
22.
Mao L, Lee DJ, Tockman MS, Erozan YS, Askin F, Sidransky D.  Microsatellite alterations as clonal markers for the detection of human cancer.   Proc Natl Acad Sci U S A.1994;91:9871-9875.
23.
Merlo A, Mabry M, Gabrielson E, Vollmer R, Baylin SB, Sidransky D. Frequent microsatellite instability in primary small cell lung cancer.  Cancer Res.1994;54:2098-2101.
24.
Yamamoto Y, Virmani AK, Wistuba II.  et al.   Loss of heterozygosity and microsatellite alterations in p53 and RB genes in adenoid cystic carcinoma of the salivary glands.   Hum Pathol.1996;27:1204-1210.
25.
Fong KM, Zimmerman PV, Smith PJ.  Microsatellite instability and other molecular abnormalities in non-small cell lung cancer.   Cancer Res.1995;55:28-30.
26.
Adachi J, Shiseki M, Okazaki T.  et al.  Microsatellite instability in primary and metastatic lung carcinomas.  Genes Chromosomes Cancer.1995;14:301-306.
27.
Orlow I, Lianes P, Lacombe L, Dalbagni G, Reuter VE, Cordon-Cardo C.  Chromosome 9 allelic losses and microsatellite alterations in human bladder tumors.   Cancer Res.1994;54:2848-2851.
28.
Miozzo M, Sozzi G, Musso K.  et al.   Microsatellite alterations in bronchial and sputum specimens of lung cancer patients.   Cancer Res.1996;56:2285-2288.
29.
Gleeson CM, Sloan JM, McGuigan JA, Ritchie AJ, Weber JL, Russell SE.  Ubiquitous somatic alterations at microsatellite alleles occur infrequently in Barrett's-associated esophageal adenocarcinoma.   Cancer Res.1996;56:259-263.
30.
Warren ST. The expanding world of trinucleotide repeats.  Science.1996;271:1374-1375.
31.
Laken SJ, Petersen GM, Gruber SB.  et al.   Familial colorectal cancer in Ashkenazim due to a hypermutable tract in APC.   Nat Genet.1997;17:79-83.
32.
Bedi GC, Westra WH, Farzadegan H, Pitha PM, Sidransky D. Microsatellite instability in primary neoplasms from HIV+ patients.  Nat Med.1995;1:65-68.
33.
Gaidano G, Pastore C, Gloghini A.  et al.   Microsatellite instability in KSHV/HHV-8 positive body-cavity-based lymphoma.   Hum Pathol.1997;28:748-750.
34.
Peterschmitt AM, Moro DG, Perron HJ, Brambilla EM. Bronchoalveolar carcinoma, lymphoid interstitial pneumonia, and retroviruses.  Clin Infect Dis.1994;19:1178-1179.
35.
Hung J, Kishimoto Y, Sugio K.  et al.   Allele-specific chromosome 3p deletions occur at an early stage in the pathogenesis of lung carcinoma.   JAMA.1995;273:558-563.
36.
Thiberville L, Payne P, Vielkinds J.  et al.   Evidence of cumulative gene losses with progression of premalignant epithelial lesions to carcinoma of the bronchus.   Cancer Res.1995;55:5133-5139.
37.
Cawkwell L, Lewis FA, Quirke P.  Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC/MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction.   Br J Cancer.1994;70:813-818.
38.
Jones MH, Nakamura Y.  Detection of loss of heterozygosity at the human TP53 locus using a dinucleotide repeat polymorphism.   Genes Chromosomes Cancer.1992;5:89-90.
39.
Toguchida J, McGee TL, Paterson JC.  et al.   Complete genomic sequence of the human retinoblastoma susceptibility gene.   Genomics.1993;17:535-543.
40.
Kishimoto Y, Sugio K, Mitsudomi T.  et al.   Allele-specific loss in chromosome 9p loci in preneoplastic lesions accompanying non-small cell lung cancers.   J Natl Cancer Inst.1995;87:1224-1229.
41.
Birch DE. Simplified hot start PCR.  Nature.1996;381:445-446.
42.
Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS. "Touchdown" PCR to circumvent spurious priming during gene amplification.  Nucleic Acids Res.1991;19:4008.
43.
Wistuba II, Montellano FD, Milchgrub S.  et al.   Deletions of chromosome 3p are frequent and early events in the pathogenesis of uterine cervical carcinoma.   Cancer Res.1997;57:3154-3158.
44.
Baay MF, Quint WG, Kuodstaal J.  et al.   Comprehensive study of several general and type-specific primer pairs for detection of human papillomavirus DNA by PCR in paraffin-embedded cervical carcinomas.   J Clin Microbiol.1996;34:745-747.
45.
Wistuba II, Lam S, Behrens C.  et al.   Molecular damage in the bronchial epithelium of current and former smokers.   J Natl Cancer Inst.1997;89:1366-1373.
46.
Kosary CL, Reiss LA, Miller BA, Hankey BF, Harras A, Edwards BK.  SEER Cancer Statistics Review, 1973-1992: Tables and Graphs, National Cancer Institute.   Bethesda, Md: National Institutes of Health; 1995. NIH publication 96-2789.
47.
Loeb LA. Microsatellite instability: marker of a mutator phenotype in cancer.  Cancer Res.1994;54:5059-5063.
48.
Imai H, Nakagama H, Komatsu K.  et al.   Minisatellite instability in severe combined immunodeficiency mouse cells.   Proc Natl Acad Sci U S A.1997;94:10817-10820.
49.
Huang J, Papadopoulos N, McKinley AJ.  et al.  APC mutations in colorectal tumors with mismatch repair deficiency.  Proc Natl Acad Sci U S A.1996;93:9049-9054.
50.
Rampino N, Yamamoto H, Ionov Y.  et al.   Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype.   Science.1997;275:967-969.
51.
Dubrova YE, Nesterov VN, Krouchinsky NG.  et al.  Human minisatellite mutation rate after the Chernobyl accident.  Nature.1996;380:683-686.
52.
Fan YJ, Sadamoto S, Ninomiya Y.  et al.   Dose-response of a radiation induction of a germline mutation at a hypervariable mouse minisatellite locus.   Int J Radiat Biol.1995;68:177-183.
53.
Sekine I, Yokose T, Ogura T.  et al.   Microsatellite instability in lung cancer patients 40 years of age or younger.   Jpn J Cancer Res.1997;88:559-563.
54.
Thomas P, De Lamballerie X, Garbe L, Douagui H, Kleisbauer JP.  Detection of human papillomavirus DNA in primary lung carcinoma by nested polymerase chain reaction.   Cell Mol Biol (Noisy-le-grand).1995;41:1093-1097.
55.
Gaidano G, Pastore C, Gloghini A.  et al.   Distribution of human herpesvirus-8 sequences throughout the spectrum of AIDS-related neoplasia.   AIDS.1996;10:941-949.
Clinical Investigation
May 20, 1998

Comparison of Molecular Changes in Lung Cancers in HIV-Positive and HIV-Indeterminate Subjects

Author Affiliations

From the Hamon Center for Therapeutic Oncology Research (Drs Wistuba, Behrens, Virmani, Minna, and Gazdar and Ms Thomas) and Departments of Pathology (Drs Milchgrub, Virmani, and Gazdar), Internal Medicine (Dr Minna), and Pharmacology (Dr Minna), University of Texas Southwestern Medical Center, Dallas; Departments of Pathology, Pontificia Universidad Catolica de Chile, Santiago (Dr Wistuba); Bellevue–New York University Medical Center, New York (Dr Jagirdar); Lenox Hill Hospital, New York, NY (Dr Ioachim); and University of Pennsylvania Medical Center, Philadelphia (Dr Litzky); and Department of Cellular Pathology, Regional Hospital Center Michallon, Grenoble, France (Dr Brambilla).

JAMA. 1998;279(19):1554-1559. doi:10.1001/jama.279.19.1554
Context.—

Context.—  Human immunodeficiency virus (HIV) infection has been associated with an increasing incidence of malignancy, and HIV-infected persons have an increased incidence of primary lung carcinoma compared with the general population.

Objective.—  To investigate the molecular changes present in HIV-associated lung tumors and compare them with those present in lung carcinomas arising in HIV-indeterminate subjects ("sporadic tumors").

Design.— Convenience sample.

Subjects.—  Archival tissues from 11 HIV-positive persons and from 35 persons of indeterminate HIV status.

Setting.— University-based medical centers and affiliated hospitals.

Main Outcome Measures.—  Analysis of frequency of loss of heterozygosity (LOH) and microsatellite alteration (MA) using polymerase chain reaction and 16 polymorphic microsatellite markers at 8 chromosomal regions frequently deleted in lung cancer. Presence of HIV and human papillomavirus (HPV) sequences.

Results.—  The overall frequency of LOH at all chromosomal regions tested and the frequencies at most of the individual regions were similar in the 2 groups. Frequency of MA present in the HIV-associated tumors (0.18) was 6-fold higher than in sporadic tumors (0.03) (P<.001). At least 1 MA was present in 10 (91%) of 11 HIV-associated tumors vs 17 (48%) of 35 sporadic tumors (P=.02). Molecular changes were independent of tumor stage and gender. HIV and HPV sequences were not detected in the HIV-associated lung carcinomas.

Conclusions.—  Microsatellite alterations, which reflect widespread genomic instability, occur at greatly increased frequency in HIV-associated lung carcinomas. Although the mechanism underlying the development of increased MAs is unknown, it may play a crucial role in the development of many HIV-associated tumors.

HUMAN immunodeficiency virus (HIV) infection has an established predisposition to certain malignancies (acquired immunodeficiency syndrome [AIDS]-defining neoplasms), including Kaposi sarcoma, primary central nervous system lymphoma, non-Hodgkin lymphoma, and cervical carcinoma.1 As the AIDS epidemic advances, the number of HIV-infected subjects developing AIDS-related neoplasms has increased, and the spectrum of malignancies is expanding. Several non–AIDS-defining cancers, including lung cancer,2 are being reported at increasing incidences in HIV-infected persons.

Although substantial progress has been made in understanding the mechanism involved in the development of AIDS-related neoplasms, the pathogenesis of these malignancies is still not fully understood. Studies on the mechanism of tumorigenesis in AIDS have focused on loss of immune surveillance and possible involvement of certain infectious agents. Epstein-Barr virus has been implicated as an agent in the development of non-Hodgkin lymphoma,3 and recently a novel human herpesvirus has been linked to a predisposition for Kaposi sarcoma.4 However, it is unknown whether HIV plays a direct role in the genesis of AIDS-related neoplasms.

Lung cancer is the most frequent cause of cancer deaths in men and women in the United States,5 and tobacco smoking is accepted as the major cause.6 Although the association of lung cancer and HIV infection is rare, more than 180 cases of lung carcinoma in HIV-positive patients have been documented in as many as 22 reports713 since the first case was described in 1984.14 Because the subject has been reviewed recently,13 only key references describing multiple cases are mentioned herein. Parker et al,13 in a recent population-based epidemiological study in Texas, identified primary lung cancers in 36 subjects with HIV infection or AIDS. Although no adjustment was made for age at onset or smoking history, the observed/expected ratio for primary lung cancer compared with that of the US population was 6.5. As with the AIDS-defining neoplasms, lung carcinomas arising in HIV-positive persons are characterized by aggressive clinical behavior with more advanced stage and shortened survival vs similar neoplasms ("sporadic cancers") in HIV-seronegative or HIV-indeterminate persons.2

Lung cancer pathogenesis is characterized by multiple molecular changes, including activation of oncogenes and loss of known and putative tumor suppressor genes (TSGs).15 Many mutations and allelic losses, especially those involving TSGs, have been described in clinically evident lung cancers.16,17 Microsatellites are highly polymorphic short tandem repeat DNA sequences. Because they are abundantly and evenly distributed throughout the genome and are easily analyzed by polymerase chain reaction (PCR)-based methods,18 they are frequently used for studies of allelic loss in tumors. The fidelity of normal microsatellite replication appears to be relatively low as microsatellite loci are highly polymorphic.19 These polymorphisms presumably arise due to unrepaired slippage during DNA replication.20

Changes in microsatellite size (both additions and deletions) are another genetic change seen in many cancers and other human diseases and are associated with at least 4 disease states: (1) In hereditary nonpolyposis colon cancer, the inherited defects in DNA mismatch repair enzymes result in large-scale genetic instability with the formation of a ladderlike pattern of microsatellites of various sizes that replaces the normal allele pattern.19,21 (2) Another form of microsatellite change, referred to as microsatellite alteration (MA) by Sidransky,22,23 ourselves,24 and others, 2529 occurs when only a single allele of altered size is found, as in many forms of sporadic cancers, including lung cancer.23,24 Microsatellite alterations usually involve dinucleotide repeats and less often, polynucleotide repeats. (3) Expansions of unstable trinucleotide repeats in the coding regions of some genes can result in familial neuromuscular disorders.30 (4) In Ashkenazi Jews, a T to A polymorphism at APC gene nucleotide 3920 results in a hypermutable tract of (A)8 sequences, indirectly causing a familial colorectal cancer predisposition.31

An increased rate of MAs has been described in 7 of 13 AIDS-associated malignancies including Kaposi sarcoma (2 of 5 cases), aggressive non-Hodgkin lymphoma (4 of 6 cases), and anal intraepithelial neoplasm (1 of 2 cases) occurring in HIV-positive patients.32 Recently, MAs have been described in a single case of AIDS-associated body cavity–based lymphoma33 although no HIV-negative control cases were examined.

In the present report, we investigated the incidence of loss of heterozygosity (LOH) and MA at several chromosomal loci frequently deleted in lung cancer in a series of 11 lung carcinomas from HIV-positive patients, and compared the results to 35 sporadic lung carcinomas occurring in HIV-indeterminate patients. We also assessed HIV and human papillomavirus (HPV) sequences.

MATERIALS AND METHODS
Study Populations and Tumor Specimens

Paraffin-embedded archival materials from 11 HIV-associated lung carcinomas were obtained from 4 institutions in the United States (10 cases) and from 1 institution in France (1 case). We selected all cases for which archival paraffin-embedded tumor, along with nontumor tissue (as a source of constitutional DNA), were available to us. The tumors represented 10 primary lung carcinomas and 1 metastasis of a lung cancer in a cervical lymph node. Further clinicopathological information is presented in Table 1. Additional details of the French case are available from a published report.34 Six of the tumor specimens were received from the AIDS Malignancy Bank, a National Institutes of Health (NIH)-sponsored resource. We compared the findings from the HIV-associated carcinomas with those from 35 sporadic primary lung carcinomas representing all major histological types from patients undergoing curative intent surgical resections. The latter consisted of 10 adenocarcinomas, 11 squamous cell carcinomas, and 14 small cell lung carcinomas. The non–small cell lung carcinoma samples were selected randomly from a large series of surgically resected archival cases available to us. Because small cell lung carcinoma tumors are seldom resected, even when apparently limited in extent, we used all resected cases available to us. Because the samples were derived from curative intent surgical resections, most sporadic tumors were early stage (stages I and II for non–small cell carcinomas and limited stage for the small cell carcinoma cases). To assess whether tumor stage influenced our observations, we selected 18 established cell lines from sporadic non–small cell lung carcinomas representing early (stages I and II [n=10])and advanced (stage IV [n=8]) clinical stages. Because cell lines from early (ie, limited) stage small cell lung carcinoma were not available, we limited the cell line study to non–small cell lung carcinoma. All available non–small cell lung carcinoma cell lines having paired lymphoblastoid cell lines (as a source of constitutional DNA) were used. Their histological types were as follows: 13 adenocarcinomas and 2 squamous cell, 1 adenosquamous, and 2 large cell carcinomas.

Microdissection and DNA Extraction

Microdissection and DNA extraction were performed as described.35 All slides were stained with hematoxylin-eosin. Microdissection and DNA extraction were performed using the noncoverslipped slides. Precisely identified areas of invasive carcinoma were microdissected under microscopic visualization without contamination with normal cells. Nonmalignant lung tissue or lymphocytes from the same sections were used as a source of constitutional DNA. After DNA extraction, 5 µL of the proteinase K–digested samples, containing DNA from at least 100 cells, was used for each multiplex PCR reaction.

Polymorphic DNA Markers and PCR for LOH and MA Analyses

To evaluate LOH and MA, we used primers flanking dinucleotide (n=11) and polynucleotide (n=5) microsatellite repeat polymorphisms located at the following genes or chromosomal regions: 3p12 (D3S1274), 3p14.2 (D3S4103 at the FHIT gene), 3p21 (D3S1029, D3S1478, D3S1447, ITIH-1, K1.CA), 3p22-24.2 (D3S2432, D3S1351, D3S1537), 5q22 (L5.71 in the APC-MCC region), 9p21 (IFNA, D9S1478), 13q14 (dinucleotide repeat at the RB gene [RB CA]), and 17p13 (TP53 dinucleotide [TP53 CA] and pentanucleotide repeats [TP53 penta]). Primer sequences can be obtained from the Genome Database [http://gdbwww.gdb.org], with 5 exceptions: ITIH-1,36 pentanucleotide,37 and dinucleotide38 repeats in the TP53 gene, dinucleotide repeat in the RB gene,39 and the K1.CA dinucleotide marker identified in our laboratory (forward primer 5‘-CATCCTCTCACCATGTACAG-3; reverse primer 5‘-CAAGGGGACAGA GACTTTG-3‘).

Nested PCR40 or 2-round PCR (using the same primers in 2 consecutive amplifications) methods were used. Multiplex PCR was done during the first amplification, followed by uniplex PCR for individual markers. In the multiplex PCR, 6 markers were amplified during the same reaction. Volumes of 50 µL were used for each multiplex reaction, containing 20-mmol/L Tris buffer (pH 8.3), 50-mmol/L potassium chloride, 2.5-mmol/L magnesium chloride, 400 µmol of each deoxynucleotide triphosphate (dATP, dCTP, dGTP, dTTP) per liter, 0.5 µmol of each forward and reverse primer per liter, and 3.5 U of AmpliTaq Gold (Perkin Elmer, Foster City, Calif).41 The first PCR product was diluted in a 1:10 ratio with double-distilled, sterile water and used for the second PCR reaction, which was performed in a 10-µL reaction volume containing 20-mmol/L Tris buffer (pH 8.3), 50-mmol/L potassium chloride, 1.5-mmol/L magnesium chloride, 100 µmol of each deoxynucleotide triphosphate (dATP, dCTP, dGTP, dTTP) per liter, 0.5 µmol of each forward and reverse primer per liter, 0.25 U of AmpliTaq polymerase (GIBCO-BRL, Alameda, Calif), 0.25 µL of dCTP tagged with phosphorus 32 (3000 Ci/mol, 10 mCi/mL, Amersham LIFE SCIENCE, Arlington Heights, Ill), and 1 to 2 µL of diluted first PCR product. For both PCR reactions, a "touchdown" PCR42 was performed. After initial denaturation at 94°C for 4 to 8 minutes, 11 cycles each consisting of denaturation at 95°C for 20 seconds, annealing at 65°C to 56°C for 55 seconds, and extension at 72°C for 20 seconds were performed. This was followed by an additional 24 cycles, which included denaturation at 90°C for 20 seconds, annealing at 55°C for 20 seconds, and extension at 72°C for 20 seconds. Then, a 5-µL aliquot of each second PCR reaction was diluted 1:4 with loading buffer, heat denatured, and separated by electrophoresis on a denaturing 6% polyacrylamide gel containing urea as published.35

For each case, constitutional heterozygosity was determined by examination of nonmalignant tissue. Loss of heterozygosity was scored by visual detection of complete absence of the 1 tumor allele in heterozygous (ie, "informative") cases. Microsatellite alterations were detected by a shift in the mobility of 1 allele (Figure 1), irrespective of whether it was accompanied by LOH. Microsatellite alterations consisted of either deletions (increased electrophoretic mobility) or insertions (decreased electrophoretic mobility).

Detection of HIV and Human Papillomavirus (HPV) Sequences

The presence of HIV sequences was assessed in the lung carcinoma–microdissected tissues by a PCR-based method using Gene-Amplifier HIV-1 reagents (Perkin-Elmer, Branchburg, NJ), utilizing the SK38 and SK39 primers and targeting a highly conserved 115–base pair segment of the HIV-1 Gag gene. The positive control was supplied by the manufacturer. This method is sensitive enough to detect 10 input copies of HIV-1–positive control DNA per reaction in paraffin-embedded tissue.

The presence of HPV sequences was assessed for using general and type-specific primers designed for paraffin-extracted DNA as previously described.43 Specific primers were used to identify high (HPV 16 and 18) and intermediate (HPV 31 and 33) oncogenic risk–HPV strains.44 DNA extracted from human cell lines CaSki (HPV 16) and HeLa (HPV 18) (obtained from the American Type Culture Collection, Rockville, Md) was used as a positive control for HPV analyses.

Data Analysis
To compare the total frequencies of LOH and MA in lung carcinomas, we have devised 2 indices,45 calculated as follows:

The fractional regional loss index indicates LOH for all informative chromosomal regions per case (maximum of 8 regions per case). In some instances, we were able to increase the number of regions that were informative by using multiple markers to analyze individual regions (when available and suitable for analysis of archival paraffin-embedded materials). If a marker for a region was informative (ie, heterozygous), the region was regarded as informative, and if 1 or more of the markers showed LOH, we regarded the region as demonstrating loss. The MA index indicates the total frequency of MA expressed as a fraction per case (n=16 markers tested). Because MAs at individual markers occur independently of chromosomal region and informativeness, data from all markers were used.

Statistical analyses were performed using the nonparametric Wilcoxon and Fisher exact tests. Probability values of P<.05 were regarded as statistically significant.

RESULTS
Clinicopathological Data for HIV-Associated and Sporadic Lung Carcinomas

As presented in Table 1, the 11 HIV-associated lung carcinomas consisted of 4 (36%) adenocarcinomas, 4 (36%) squamous cell carcinomas, 2 (18%) small cell carcinomas, and 1 (9%) large cell carcinoma. Seven of these tumors were surgically resected, and the rest were obtained by bronchoscopy (n=2) or at autopsy (n=2). We compared the findings from the HIV-associated carcinomas with those from 35 sporadic primary lung carcinomas from patients undergoing curative intent surgical resections.

All of the HIV-associated lung carcinoma patients were male and relatively young, with a median age of 45 years (range, 26-60 years). All 10 patients from whom data were available were current or former smokers (median exposure, 20 pack-years). Tumor stage varied from early (6 patients, stages I and II) to late (4 patients, stages III and IV), and, for the small cell lung carcinoma patients, extensive. In 6 of 7 patients from whom information was available, HIV seropositivity was noted at the time of tumor diagnosis, while in the remaining patient, HIV seropositivity was detected 4 years prior to tumor diagnosis. Of the 8 patients from whom information was available, 5 had AIDS as defined by CD4 counts below 0.20×109 /L, and 1 of these had received therapy with zidovudine for 4 years prior to tumor diagnosis.

The sporadic lung cancer patient group showed a nearly equal gender distribution. All but 1 were heavy smokers with a median exposure of 37 pack-years (range, 0-120). The 1 nonsmoker was a woman with a long history of heavy passive exposure to cigarette smoke. Their median age was 61 years (range, 40-75 years), which is closer to that of persons with lung cancer in the general US population, for whom the median age at lung cancer diagnosis is 68 years.46

LOH Frequency in Lung Carcinomas

Analyses were performed on tumor samples that had been carefully microdissected and LOH was detected by complete absence of 1 of the 2 alleles present in constitutional DNA of informative cases (Figure 1). Frequencies of LOH and the pattern of allelic losses were similar in HIV-associated and sporadic lung carcinomas (Figure 2, A, Table 2). Almost identical relatively high frequencies of total LOH at the 8 chromosomal regions analyzed were present in the 2 groups, as represented by the fractional regional loss indices (0.67 and 0.68, respectively) (Figure 2, A). In addition, the frequencies at individual regions were similar in the 2 groups with high frequencies of allelic loss at chromosomal regions 3p12, 3p21, 3p22-24.2, 9p21, 13q (RB gene), and 17p (TP53 gene) (Table 2). Differences were noted at 2 sites: allelic loss frequencies were higher at 5q22 (APC/MCC gene region) in HIV-associated than in sporadic carcinomas (86% vs 39%; P=.007) but were lower at the FHIT gene (25% vs 77%; P=.03, by the Fisher exact test).

MA Frequency in Lung Carcinomas

Although there was considerable variability, the mean MA index of the HIV-associated tumors (0.18) was 6-fold greater than the mean index of the sporadic cases (0.03) (Figure 2, B) and the difference was highly significant (P<.001, Wilcoxon test). Examples of various MA patterns are presented in Figure 1. Because artifacts resulting from PCR amplification may be mistaken for MAs, especially when minute amounts of input DNA are utilized, all examples of MAs were confirmed using DNA microdissected from a replicate microsection. Using 16 polymorphic markers, at least 1 MA was identified in 10 (91%) of 11 HIV-associated tumors vs 17 (48%) of 35 sporadic tumors (P=.02). The patient who received zidovudine for 4 years prior to the diagnosis of lung cancer had a relatively low MA index (0.06).

Of the MAs identified, insertions (61% and 50%, respectively) and deletions (39% and 50%, respectively) occurred at similar frequencies in HIV-associated and sporadic tumors. We determined the frequency at which both LOH and MA occurred at individual informative loci. In both HIV-associated and sporadic tumors, the frequencies were similar (20% and 24%, respectively) (P>.05). For 12 (75%) of the 16 markers, the rates were higher in the HIV-associated group (up to 9-fold higher), while for the other 4 markers the rates were either similar (n=2) or slightly higher in the sporadic group (n=2) (Figure 3). Although no obvious differences were detected in the MA frequencies between dinucleotide or polynucleotide microsatellite repeat markers (Figure 3), the frequencies varied considerably for individual markers.

Because all the HIV-associated lung carcinoma patients were male, we also compared their MA index (0.18) with that of cases of sporadic lung tumors (0.04) arising only in males (n=20), and the differences were still significant (P=.004). To exclude the possibility that tumor stage influenced the frequency of MA, we determined that the MA index of 10 non–small cell lung carcinoma cell lines from early stage cases (stages I and II, mean MA index = 0.03) was similar to that from 8 cases of late-stage tumors (stage IV, mean MA index = 0.02, P>.05).

MA in Nonmalignant Tissues

In 2 HIV-associated cases, multiple samples of nonmalignant tissue were available for analysis (Table 3). Of these cases, 1 tumor had a relatively high MA index (0.38), while the other had a low index (0.06). In 15 samples of nonmalignant tissue from these 2 cases, no MAs were detected.

Viral Sequences in HIV-Associated Lung Carcinomas

Human immunodeficiency virus or HPV sequences were not detected in any of the microdissected lung carcinomas from the HIV-positive persons.

COMMENT

Because no information is currently available about the molecular changes involved in lung carcinoma arising in HIV-positive patients, we investigated the frequencies of LOH and MA present in 11 lung carcinomas from 11 HIV-positive persons and compared them with those from sporadic lung carcinomas of 35 persons. After careful microdissection of the tumors, extracted DNA was analyzed by PCR for LOH and MA at multiple chromosomal regions frequently deleted in sporadic lung cancer.

Reports of HIV-associated lung carcinomas suggest that the natural history of the disease is different in persons with lung cancer in the general population.2,7,13 Lung carcinomas arising in HIV-positive patients are characterized by young age at diagnosis (median, 38 years), occurring almost exclusively in males, slightly increased frequency of adenocarcinoma histology, late stage at presentation, and shortened survival.7 Although cigarette smoking appears to be as prevalent among HIV-positive lung cancer patients as in other lung cancer patients, the former usually have a lesser history of smoking exposure (usually <20 pack-years).7 Our cohort of 11 subjects with HIV-associated lung cancer fits this profile. In contrast, the profile of our 35 cases with sporadic lung cancer more closely resembled that of lung cancer arising in the general population.

Many mutations, especially those involving TSGs, have been described in clinically evident lung cancers.16,17 We determined the frequencies of LOH and MA at 8 chromosomal regions frequently deleted in lung tumors, utilizing 16 microsatellite markers for analysis of HIV-associated and sporadic lung carcinoma. Of interest, the overall frequency of LOH at all regions and the frequencies at most individual regions were similar in the 2 groups, with high frequencies of allelic loss at most chromosomal regions, although significant differences were noted at 2 sites: allelic loss frequencies were higher at 5q22 (APC/MCC gene region) in HIV-associated than in sporadic carcinoma and were lower at 3p14.2 (FHIT gene). However, because of the small sample size, we interpret the overall pattern of allelic losses in the 2 groups as similar.

A significantly higher frequency of MA was present in the HIV-associated tumors. At least 1 MA was present in 10 (91%) of 11 HIV-associated tumors vs 17 (48%) of 35 sporadic tumors. These differences were independent of gender and tumor stage. Although the frequencies varied considerably for individual markers, there were no obvious differences in the MA frequencies between dinucleotide and polynucleotide microsatellite repeat markers. This finding is somewhat surprising because the frequency of MA detected by polynucleotide-repeat markers has been reported to be higher than that detected by dinucleotide markers.22 Artifacts resulting from PCR amplification may be mistaken for MAs; thus, all apparent MAs were confirmed utilizing DNA from material derived from a replicate microsection. Microsatellite alterations resulting from deletions or insertions occurred at similar frequencies in HIV-associated and sporadic tumors. In HIV-associated tumors, there was no increased frequency of allelic loss at loci demonstrating MAs.

Microsatellite alterations have been described in a variety of cancer types. While the relationship of MAs to DNA repair mechanisms has not been established, MAs probably represent evidence of some form of genomic instability.47 Most microsatellite sequences (and their alterations) arise in noncoding regions of the genome and are not in the direct pathway of tumorigenesis. Although long stretches of microsatellite sequences are rare in coding regions, another form of nucleotide repeat, known as minisatellites, is more frequent.48 Minisatellites are tandem repeat arrays that are locus specific but have a consensus sequence that varies in the range of 5 to 100 base pairs.48 In patients with hereditary nonpolyposis colon cancer, size changes affecting minisatellites present in known or putative genes, including the APC gene, may result in frameshift mutations.49,50 Thus, it is possible that MAs also may result in mutations affecting key genes involved in cancer pathogenesis. Microsatellite alterations have also been described in other HIV-associated tumors and may be involved in their pathogenesis.32,33

Although there are several theories, the mechanism by which increased rates of MA occur in HIV-associated tumors is not known. While increased rates occur (probably in all somatic tissues) after radiation exposure in humans51 and experimental animals,52 none of our HIV-positive subjects had received prior radiotherapy. The possibility also exists that zidovudine, a thymidine analog, could interfere with the DNA repair mechanism. However, only 1 of our patients with HIV infection had received zidovudine prior to the diagnosis of lung cancer, and the MA index of his tumor was relatively low. A relatively high frequency of MA has been described in lung cancers arising in young nonsmoking subjects,53 suggesting a genetic predisposition. Microsatellite alterations are not due to direct effects of HIV viral integration into the genome of affected cells, as HIV sequences were absent in all of the HIV-associated lung carcinomas, a finding also reported for most other HIV-associated tumors (reviewed by Bedi et al32). Our failure to identify MAs in nonmalignant tissues diminishes the possibility that HIV induces a general somatic effect on the mutation repair system throughout the body. Human papillomavirus sequences, which may be present in HIV-associated cervical and anogenital carcinomas (and in a subset of sporadic lung carcinomas),54 were absent in the HIV-associated lung carcinoma. We cannot exclude a direct role of other viruses described in HIV-associated tumors including human herpesvirus 8 frequency.55 Immunosuppressive states may result in minisatellite instability.48 Thus, immunosuppression, which develops in most HIV-infected subjects, may play a role in tumor development via multiple mechanisms.

Our findings suggest that widespread genomic instability, as manifested by the increased frequency of MA, occurs frequently in HIV-associated lung carcinoma. We did not detect MA in nonmalignant tissue from 2 HIV-positive cases and it remains to be determined at what stage MAs develop during the multistage pathogenesis of lung cancer. If they develop relatively early, during the preneoplastic process, determination of MA frequency may provide a useful method for cancer risk assessment in HIV-positive persons. Although the mechanism underlying the development of increased numbers of MAs is unknown, it may play a crucial role in the development of many HIV-associated tumors.

References
1.
Biggar RJ, Rabkin CS. The epidemiology of AIDS-related neoplasms.  Hematol Oncol Clin North Am.1996;10:997-1010.
2.
Remick SC. Non–AIDS-defining cancers.  Hematol Oncol Clin North Am.1996;10:1203-1213.
3.
Knowles DM. Etiology and pathogenesis of AIDS-related non-Hodgkin's lymphoma.  Hematol Oncol Clin North Am.1996;10:1081-1111.
4.
Miles SA. Pathogenesis of AIDS-related Kaposi's sarcoma.  Hematol Oncol Clin North Am.1996;10:1011-1021.
5.
Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1997.  CA Cancer J Clin.1997;47:5-27.
6.
Parkin DM, Pisani P, Lopez AD, Masuyer E.  At least one in seven cases of cancer is caused by smoking: global estimates for 1985.   Int J Cancer.1994;59:494-504.
7.
Fraire AE, Awe RJ. Lung cancer in association with human immunodeficiency virus infection.  Cancer.1992;70:432-436.
8.
Sridhar KS, Flores MR, Raub WJ, Saldana M.  Lung cancer in patients with human immunodeficiency virus infection compared with historic control subjects.   Chest.1992;102:1704-1708.
9.
Karp J, Profeta G, Marantz PR, Karpel JP. Lung cancer in patients with immunodeficiency syndrome.  Chest.1993;103:410-413.
10.
Vaccher E, Tirelli U, Spina M.  et al.   Lung cancer in 19 patients with HIV infection: the Italian Cooperative Study Group on AIDS and Tumors (GICAT).   Ann Oncol.1993;4:85-86.
11.
Fishman JE, Schwartz DS, Sais GJ, Flores MR, Sridhar KS.  Bronchogenic carcinoma in HIV-positive patients: findings on chest radiographs and CT scans.   AJR Am J Roentgenol.1995;164:57-61.
12.
White CS, Haramati LB, Elder KH, Karp J, Belani CP.  Carcinoma of the lung in HIV-positive patients: findings on chest radiographs and CT scans.   AJR Am J Roentgenol.1995;164:593-597.
13.
Parker MS, Leveno DM, Campbell TJ, Worrell JA, Carozza SE. AIDS related bronchogenic carcinoma: fact or fiction?  Chest.1998;113:154-161.
14.
Irwin LE, Begandy MK, Moore TM.  Adenosquamous carcinoma of the lung and the acquired immunodeficiency syndrome.   Ann Intern Med.1984;100:158.
15.
Knudson AG. Hereditary cancers: clues to mechanisms of carcinogenesis.  Br J Cancer.1989;59:661-666.
16.
Gazdar AF, Bader S, Hung J.  et al.   Molecular genetic changes found in human lung cancer and its precursor lesions.   Cold Spring Harb Symp Quant Biol.1994;59:565-572.
17.
Minna JD, Sekido Y, Fong K, Gazdar AF. Molecular biology of lung cancer.  In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer: Principles and Practice of Oncology. 5th ed. Philadelphia, Pa: JB Lippincott; 1997:849-857.
18.
Weber JL, May PE.  Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction.   Am J Hum Genet.1989;44:388-396.
19.
Shibata D. Molecular tumor clocks and dynamic phenotype.  Am J Pathol.1997;151:643-646.
20.
Strand M, Prolla TA, Liskay RM, Petes TD.  Destabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair.   Nature.1993;365:274-276.
21.
Liu B, Parsons R, Papadopoulos N.  et al.   Analysis of mismatch repair genes in hereditary non-polyposis colorectal cancer patients.   Nat Med.1996;2:169-174.
22.
Mao L, Lee DJ, Tockman MS, Erozan YS, Askin F, Sidransky D.  Microsatellite alterations as clonal markers for the detection of human cancer.   Proc Natl Acad Sci U S A.1994;91:9871-9875.
23.
Merlo A, Mabry M, Gabrielson E, Vollmer R, Baylin SB, Sidransky D. Frequent microsatellite instability in primary small cell lung cancer.  Cancer Res.1994;54:2098-2101.
24.
Yamamoto Y, Virmani AK, Wistuba II.  et al.   Loss of heterozygosity and microsatellite alterations in p53 and RB genes in adenoid cystic carcinoma of the salivary glands.   Hum Pathol.1996;27:1204-1210.
25.
Fong KM, Zimmerman PV, Smith PJ.  Microsatellite instability and other molecular abnormalities in non-small cell lung cancer.   Cancer Res.1995;55:28-30.
26.
Adachi J, Shiseki M, Okazaki T.  et al.  Microsatellite instability in primary and metastatic lung carcinomas.  Genes Chromosomes Cancer.1995;14:301-306.
27.
Orlow I, Lianes P, Lacombe L, Dalbagni G, Reuter VE, Cordon-Cardo C.  Chromosome 9 allelic losses and microsatellite alterations in human bladder tumors.   Cancer Res.1994;54:2848-2851.
28.
Miozzo M, Sozzi G, Musso K.  et al.   Microsatellite alterations in bronchial and sputum specimens of lung cancer patients.   Cancer Res.1996;56:2285-2288.
29.
Gleeson CM, Sloan JM, McGuigan JA, Ritchie AJ, Weber JL, Russell SE.  Ubiquitous somatic alterations at microsatellite alleles occur infrequently in Barrett's-associated esophageal adenocarcinoma.   Cancer Res.1996;56:259-263.
30.
Warren ST. The expanding world of trinucleotide repeats.  Science.1996;271:1374-1375.
31.
Laken SJ, Petersen GM, Gruber SB.  et al.   Familial colorectal cancer in Ashkenazim due to a hypermutable tract in APC.   Nat Genet.1997;17:79-83.
32.
Bedi GC, Westra WH, Farzadegan H, Pitha PM, Sidransky D. Microsatellite instability in primary neoplasms from HIV+ patients.  Nat Med.1995;1:65-68.
33.
Gaidano G, Pastore C, Gloghini A.  et al.   Microsatellite instability in KSHV/HHV-8 positive body-cavity-based lymphoma.   Hum Pathol.1997;28:748-750.
34.
Peterschmitt AM, Moro DG, Perron HJ, Brambilla EM. Bronchoalveolar carcinoma, lymphoid interstitial pneumonia, and retroviruses.  Clin Infect Dis.1994;19:1178-1179.
35.
Hung J, Kishimoto Y, Sugio K.  et al.   Allele-specific chromosome 3p deletions occur at an early stage in the pathogenesis of lung carcinoma.   JAMA.1995;273:558-563.
36.
Thiberville L, Payne P, Vielkinds J.  et al.   Evidence of cumulative gene losses with progression of premalignant epithelial lesions to carcinoma of the bronchus.   Cancer Res.1995;55:5133-5139.
37.
Cawkwell L, Lewis FA, Quirke P.  Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC/MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction.   Br J Cancer.1994;70:813-818.
38.
Jones MH, Nakamura Y.  Detection of loss of heterozygosity at the human TP53 locus using a dinucleotide repeat polymorphism.   Genes Chromosomes Cancer.1992;5:89-90.
39.
Toguchida J, McGee TL, Paterson JC.  et al.   Complete genomic sequence of the human retinoblastoma susceptibility gene.   Genomics.1993;17:535-543.
40.
Kishimoto Y, Sugio K, Mitsudomi T.  et al.   Allele-specific loss in chromosome 9p loci in preneoplastic lesions accompanying non-small cell lung cancers.   J Natl Cancer Inst.1995;87:1224-1229.
41.
Birch DE. Simplified hot start PCR.  Nature.1996;381:445-446.
42.
Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS. "Touchdown" PCR to circumvent spurious priming during gene amplification.  Nucleic Acids Res.1991;19:4008.
43.
Wistuba II, Montellano FD, Milchgrub S.  et al.   Deletions of chromosome 3p are frequent and early events in the pathogenesis of uterine cervical carcinoma.   Cancer Res.1997;57:3154-3158.
44.
Baay MF, Quint WG, Kuodstaal J.  et al.   Comprehensive study of several general and type-specific primer pairs for detection of human papillomavirus DNA by PCR in paraffin-embedded cervical carcinomas.   J Clin Microbiol.1996;34:745-747.
45.
Wistuba II, Lam S, Behrens C.  et al.   Molecular damage in the bronchial epithelium of current and former smokers.   J Natl Cancer Inst.1997;89:1366-1373.
46.
Kosary CL, Reiss LA, Miller BA, Hankey BF, Harras A, Edwards BK.  SEER Cancer Statistics Review, 1973-1992: Tables and Graphs, National Cancer Institute.   Bethesda, Md: National Institutes of Health; 1995. NIH publication 96-2789.
47.
Loeb LA. Microsatellite instability: marker of a mutator phenotype in cancer.  Cancer Res.1994;54:5059-5063.
48.
Imai H, Nakagama H, Komatsu K.  et al.   Minisatellite instability in severe combined immunodeficiency mouse cells.   Proc Natl Acad Sci U S A.1997;94:10817-10820.
49.
Huang J, Papadopoulos N, McKinley AJ.  et al.  APC mutations in colorectal tumors with mismatch repair deficiency.  Proc Natl Acad Sci U S A.1996;93:9049-9054.
50.
Rampino N, Yamamoto H, Ionov Y.  et al.   Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype.   Science.1997;275:967-969.
51.
Dubrova YE, Nesterov VN, Krouchinsky NG.  et al.  Human minisatellite mutation rate after the Chernobyl accident.  Nature.1996;380:683-686.
52.
Fan YJ, Sadamoto S, Ninomiya Y.  et al.   Dose-response of a radiation induction of a germline mutation at a hypervariable mouse minisatellite locus.   Int J Radiat Biol.1995;68:177-183.
53.
Sekine I, Yokose T, Ogura T.  et al.   Microsatellite instability in lung cancer patients 40 years of age or younger.   Jpn J Cancer Res.1997;88:559-563.
54.
Thomas P, De Lamballerie X, Garbe L, Douagui H, Kleisbauer JP.  Detection of human papillomavirus DNA in primary lung carcinoma by nested polymerase chain reaction.   Cell Mol Biol (Noisy-le-grand).1995;41:1093-1097.
55.
Gaidano G, Pastore C, Gloghini A.  et al.   Distribution of human herpesvirus-8 sequences throughout the spectrum of AIDS-related neoplasia.   AIDS.1996;10:941-949.
×