A perplexing array of immune abnormalities have been described in cutaneous T-cell lymphoma. This article discusses the mechanisms underlying these immune abnormalities. The immune processes that have become disordered are examined with reference to the steps involved in generating the disease and how these steps can be corrected therapeutically.
Pseudoepitheliomatous hyperplasia (PEH) in biopsies of CD30+ anaplastic large-cell lymphoma (ALCL) is present infrequently and is of unknown prognostic value and significance. Our goal was to review the clinicopathologic features of cases of ALCL with PEH, study their course, and review the literature on the subject. Biopsy specimens of all cases of CD30+ lymphoproliferative disorders (59) were retrieved from the files of Yale Dermatopathology Laboratory over a 17-year period and reviewed. We identified 4 cases of ALCL (7%) exhibiting prominent PEH. All 4 patients presented with 1 or 2 nodules. In 2 patients, the lesions spontaneously regressed within a few months after initial diagnosis. One patient chose to have an excision in which only a small number of CD30+ cells were present. We were unable to obtain follow-up for the fourth patient. In the spectrum of CD30+ lymphoproliferative disorders, cases of ALCL with PEH are infrequent. In addition to the 4 cases described here, in our review of the literature we found 35 cases of ALCL with PEH. Most of these patients present with 1 or a few lesions. In the majority of these cases, the lesions started showing evidence of clinical spontaneous regression and even complete resolution within a few months of initial diagnosis. The clinicopathologic correlation between ALCL and PEH has not been emphasized. Because most of these cases follow a relatively benign clinical course, we recommend a more conservative approach in the clinical management of these patients.
Annular epidermolytic ichthyosis is a distinct phenotypic variant of bullous congenital ichthyosiform erythroderma that has recently been described in two separate kindreds. Individuals with this variant present with bullous ichthyosis in early childhood and hyperkeratotic lichenified plaques in the flexural areas and extensor surfaces at later ages. Characteristically, they also develop intermittent bouts of annular and polycyclic, erythematous, scaly plaques on the trunk and proximal extremities. We now describe a third kindred with annular epidermolytic ichthyosis. Molecular analysis of this family revealed a novel mutation resulting in an isoleucine to threonine substitution at residue 107 (codon 446) within the highly conserved helix termination motif at the end of the rod domain of keratin 10. Annular epidermolytic ichthyosis is a distinct phenotypic variant of bullous congenital ichthyosiform erythroderma that has recently been described in two separate kindreds. Individuals with this variant present with bullous ichthyosis in early childhood and hyperkeratotic lichenified plaques in the flexural areas and extensor surfaces at later ages. Characteristically, they also develop intermittent bouts of annular and polycyclic, erythematous, scaly plaques on the trunk and proximal extremities. We now describe a third kindred with annular epidermolytic ichthyosis. Molecular analysis of this family revealed a novel mutation resulting in an isoleucine to threonine substitution at residue 107 (codon 446) within the highly conserved helix termination motif at the end of the rod domain of keratin 10. annular epidermolytic ichthyosis bullous congenital ichthyosiform erythroderma ichthyosis bullosa of Siemens Bullous congenital ichthyosiform erythroderma (BCIE) is an autosomal dominant disorder manifest by widespread blistering and an ichthyotic erythroderma at birth that persists into adulthood. Histologically there is diffuse epidermolytic degeneration in the lower spinous layer of the epidermis. This disorder is caused by mutations in either keratin 1 or 10 (Kl or Kl0) (Cheng et al., 1992Cheng J. Syder A.J. Yu Q.C. Letai A. Paller A.S. Fuchs E. The genetic basis of epidermolytic hyperkeratosis: a disorder of differentiation specific epidermal keratin genes.Cell. 1992; 70: 811-819Abstract Full Text PDF PubMed Scopus (273) Google Scholar;Chipev et al., 1992Chipev C.C. Korge B.P. Markova N. Bale S.J. DiGiovanna J.J. Compton J.G. Steinert P.M. A leucine-proline mutation in the H1 subdomain of keratin 1 causes epidermolytic hyperkeratosis.Cell. 1992; 70: 821-828Abstract Full Text PDF PubMed Scopus (244) Google Scholar;Rothnagel et al., 1992Rothnagel J.A. Dominey A.M. Dempsey L.D. et al.Mutations in the rod domains of keratin 1 and 10 in epidermolytic hyperkeratosis.Science. 1992; 257: 1128-1130Crossref PubMed Scopus (305) Google Scholar). Ichthyosis bullosa of Siemens (IBS) is an autosomal dominant disorder characterized by a more benign course in which affected individuals are often born with diffuse erythema and blistering that becomes more localized and actually improves with age. In addition to the hyperkeratotic, lichenified flexural plaques, these individuals also exhibit a localized Mauserung or "molting effect." Histologically, the hyperkeratotic plaques reveal a more superficial epidermolytic hyperkeratosis in the upper spinous and granular layers of the epidermis. IBS is caused by mutations in the keratin 2e gene (K2e) (Kremer et al., 1994Kremer H. Zeeuwen P. McLean W.H. et al.Ichthyosis Bullosa of Siemens is caused by mutations in the Keratin 2e gene.J Invest Dermatol. 1994; 103: 286-289Crossref PubMed Scopus (86) Google Scholar;McLean et al., 1994McLean W.H. Morley S.M. Lane E.B. et al.Ichthyosis bullosa of Siemens – a disease involving keratin 2e.J Invest Dermatol. 1994; 103: 277-281Crossref PubMed Scopus (92) Google Scholar;Rothnagel et al., 1994Rothnagel J.A. Traupe H. Wojcik S. et al.Mutations in the rod domain of keratin 2e in patients with ichthyosis bullosa of Siemens.Nature Genet. 1994; 7: 485-490Crossref PubMed Scopus (124) Google Scholar).Sahn et al., 1992Sahn E.E. Weimer C.E. Garen P.D. Annular epidermolytic ichthyosis: a unique phenotype.J Am Acad Dermatol. 1992; 27: 348-355Abstract Full Text PDF PubMed Scopus (26) Google Scholar reported a new clinical phenotype that resembles clinical and histologic features of both BCIE and IBS. They called this phenotype annular epidermolytic ichthyosis (AEI). Recently, we described a second kindred with clinical and histologic features similar to AEI (Joh et al., 1997Joh G.Y. Traupe H. Metze D. et al.A novel dinucleotide mutation in keratin 10 in the annular epidermolytic ichthyosis variant of bullous congenital ichthyosiform erythroderma.J Invest Dermatol. 1997; 108: 357-361Abstract Full Text PDF PubMed Scopus (43) Google Scholar). Molecular analysis revealed a novel mutation in the K10 gene (a dinucleotide mutation within the 2B segment of the rod domain) in affected family members. Therefore, we proposed that AEI should be considered a variant of BCIE (Joh et al., 1997Joh G.Y. Traupe H. Metze D. et al.A novel dinucleotide mutation in keratin 10 in the annular epidermolytic ichthyosis variant of bullous congenital ichthyosiform erythroderma.J Invest Dermatol. 1997; 108: 357-361Abstract Full Text PDF PubMed Scopus (43) Google Scholar). We now report a third kindred with similar clinical and histologic findings as the other two families described as having AEI. Molecular analysis of DNA from affected and unaffected family members in our kindred revealed yet another novel mutation within the helix termination motif of K10. The proband is an 11 y old male (individual IV.2 inFigure 1) , born to nonconsanguinous parents, who was referred to us for evaluation and treatment of severe atopic dermatitis. He had no skin abnormalities at birth. At the age of 7 mo he was noted to have very dry skin, especially in the flexural areas and over the extensor surfaces. He never had frank blistering or erythroderma. Over time, the plaques in the flexural areas and over the extensor surfaces expanded centrifugally and developed marked central hyperkeratotic lichenification. He also had intermittent episodes of erythematous, scaly, annular and serpiginous plaques that would develop over the proximal extremities and trunk (Figure 2a). Hair, nails, mucous membranes, and palms and soles were normal. Thus, the clinical features are compatible with a diagnosis of AEI.Figure 2Clinical presentation and histopathology of the proband. (a) Typical flexural presentation, showing the hyperkeratotic lichenified plaques with verruciform columnar scales in linear array over the antecubital fossae, with surrounding, concentric, scaly, annular, and polycyclic plaques. (b) Histopathology of hyperkeratotic lichenified flexural plaque. Hyperkeratosis, acanthosis, and thickened granular layer. Keratinocytes in upper spinous and granular layers of epidermis demonstrate cytoplasmic vacuolization and prominent keratohyaline granules.Scale bar: 50 μm.View Large Image Figure ViewerDownload (PPT) His mother (III.3) and younger sister (IV.1) both had mild hyperkeratotic plaques over the knees and elbows, but no other cutaneous findings. His father (III.4) was unaffected. His maternal uncle (III.1) and maternal grandfather (II.4) had lesions of similar clinical appearance and severity as our proband, but both were deceased (Figure 1). A biopsy was performed from the edge of a flexural, hyperkeratotic, lichenified plaque and blood was obtained for DNA analysis. Genomic DNA was isolated from whole blood of the proband and the family members as described previously (Rothnagel et al., 1992Rothnagel J.A. Dominey A.M. Dempsey L.D. et al.Mutations in the rod domains of keratin 1 and 10 in epidermolytic hyperkeratosis.Science. 1992; 257: 1128-1130Crossref PubMed Scopus (305) Google Scholar). The 2B region of K10 was amplified from genomic DNA using specific oligonucleotides (5′-AAGTGCTGAAACTGAAATGGTGCC-3′ and 5′-TCTACCCTCTCTCCTCCCTTCCTC-3′) corresponding to bp 4373–4396 and to 4837–4860 of the published sequence (Rieger and Franke, 1988Rieger M. Franke W.W. Identification of an orthologous mammalian cytokeratin gene.J Mol Biol. 1988; 204: 841-856Crossref PubMed Scopus (89) Google Scholar). Polymerase chain reaction (PCR) conditions were 1 min at 94°C, 2 min at 55°C, and 3 min at 72°C, for 30 cycles. PCR products were purified using a PCR Purification Kit (Qiagen, Santa Clarita, CA) and were directly sequenced using the ABI PRISM Dye Terminator Cycle Sequencing Kit with AmpliTaq DNA Polymerase FS (Perkin Elmer-ABI, Foster City, CA). The sequences were run on an ABI Prism 377 DNA Sequencer (Perkin Elmer-ABI). The sequencing primer was 5′-ATAAGCGTCACCATACTC-3′, corresponding to bp 4463–4480 of the published sequence (Rieger and Franke, 1988Rieger M. Franke W.W. Identification of an orthologous mammalian cytokeratin gene.J Mol Biol. 1988; 204: 841-856Crossref PubMed Scopus (89) Google Scholar). The mutation destroys an Apo I site (5′-PuAATTPy-3′) in the 2B region of K10 in the mutant allele. Therefore, segregation of the mutation was confirmed by RFLP analysis with Apo I. The 2B region of K10 was amplified by PCR with specific oligonucleotides and digested with Apo I (New England Biolabs, Beverly, MA) at 50°C overnight. Histopathologic examination of a skin biopsy specimen from a flexural, hyperkeratotic, lichenified plaque revealed orthokeratosis, acanthosis, papillomatosis, and a thickened granular layer (Figure 2b). Also prominent were coarse keratohyaline granules, blurring of the cell boundaries, and vacuolar degeneration of the keratinocytes in the superficial spinous and granular layers, reminiscent of those findings in IBS. Because the histologic analysis of the proband was most consistent with a diagnosis of IBS, we initially screened the K2e gene for mutations. The failure to find obvious mutations in the K2e gene prompted us to examine both the K1 and the K10 genes. Direct automated sequence analysis revealed a T to C transition at position 107 (codon 446) within the 2B segment of the rod domain of K10, which converted an isoleucine residue (ATT) to a threonine (ACT) [I107T], indicated by thearrow inFigure 3(a). This mutation was found in all affected family members examined and was absent in unaffected members (Figure 4). The I107T mutation abolishes an Apo I site (5′-PuAATTPy-3′) in the mutant allele. To further confirm that this mutation was not a polymorphism and was indeed associated with the disease phenotype, 50 normal controls were screened by RFLP analysis, as well as affected and unaffected family members. A 490 bp fragment containing the 2B region of K10 was amplified by PCR and digested with Apo I. The normal allele was digested into 330 and 160 bp fragments, but the 490 bp fragment from the mutant allele was resistant to digestion. All affected individuals in this family were heterozygous for the mutant and the wild-type allele, and produced three (490, 330, and 160 bp) fragments in this assay. Unaffected individuals were homozygous for the wild-type allele, and produced only two (330 and 160 bp) fragments. This mutation was not detected in the 50 normal DNA analyzed (data not shown). Thus, the Apo I RFLP cosegregates with expression of the disease phenotype, in agreement with the autosomal dominant inheritance pattern of this disease. In this study, we describe a novel mutation in the BCIE variant, AEI, which is the first reported substitution of this highly conserved isoleucine residue (I107T) within the helix termination motif of a type I keratin gene. Although we have not presented functional data documenting that this mutation is pathogenic, several lines of evidence suggest that it is. First, this mutation was only detected in family members with clinical symptoms, and not in unaffected family members. Second, this mutation was not observed in any of the 100 normal K10 alleles that were screened in this study and there are no published reports of silent polymorphisms at this site. Third, this isoleucine residue is absolutely conserved in all type I and type II keratins sequenced to date, as well as in several other intermediate filament proteins (Hatzfeld and Weber, 1991Hatzfeld M. Weber K. Modulation of keratin intermediate filament assembly by single amino acid exchanges in the consensus sequence at the C-terminal of the rod domain.J Cell Sci. 1991; 99: 351-362PubMed Google Scholar;Irvine et al., 1997Irvine A.D. McKenna K.E. Bingham A. Nevin N.C. Hughes Ä. A novel mutation in the helix termination peptide of keratin 5 causing epidermolysis bullosa simplex Dowling-Meara.J Invest Dermatol. 1997; 109: 815-816Abstract Full Text PDF PubMed Scopus (19) Google Scholar). Cross-linking experiments predict important regions of overlap between the helix initiation and termination motifs of neighboring keratin molecules (Steinert et al., 1993Steinert P.M. Marekov L.N. Fraser Rdb Parry DaD Keratin intermediate filament structure: closslinking studies yield quantitative information on molecular dimensions and mechanism of assembly.J Mol Biol. 1993; 230: 436-452Crossref PubMed Scopus (230) Google Scholar); however, to date, the overwhelming majority of the K10 mutations identified in BCIE occur within the helix initiation motif (Rothnagel and Roop, 1995Rothnagel J.A. Roop D.R. Analysis, diagnosis, and molecular genetics of keratin disorders.Curr Opinion Dermatol. 1995: 211-218Google Scholar;Corden and McLean, 1996Corden L.D. McLean WhI Human keratin disease: Hereditary fragility of specific epithelial tissues.Exp Dermatol. 1996; 5: 297-307Crossref PubMed Scopus (186) Google Scholar). In fact, due to the lack of reported mutations in the helix termination motif of type I keratins, we previously speculated that substitutions within this region might be well tolerated and fail to produce an obvious phenotype (Rothnagel et al., 1994Rothnagel J.A. Traupe H. Wojcik S. et al.Mutations in the rod domain of keratin 2e in patients with ichthyosis bullosa of Siemens.Nature Genet. 1994; 7: 485-490Crossref PubMed Scopus (124) Google Scholar). This study clearly documents that mutations in the helix termination motif of K10 can be pathogenic; however, based on a survey of reported cases of BCIE, it is also apparent that substitutions at this particular residue of K10 are not as disruptive as those occurring in the helix initiation motif of K10, or those occurring in either the helix initiation or the termination motifs of K1 (Rothnagel and Roop, 1995Rothnagel J.A. Roop D.R. Analysis, diagnosis, and molecular genetics of keratin disorders.Curr Opinion Dermatol. 1995: 211-218Google Scholar;Corden and McLean, 1996Corden L.D. McLean WhI Human keratin disease: Hereditary fragility of specific epithelial tissues.Exp Dermatol. 1996; 5: 297-307Crossref PubMed Scopus (186) Google Scholar). Prior to this study, only three mutations had been identified in the 2B segment of K10, and these occurred within residues located just upstream of the helix termination motif.Chipev et al., 1994Chipev C.C. Yang J.M. DiGiovanna J.J. Steinert P.M. Marekov L. Compton J.G. Bale S.J. Preferential sites in keratin 10 that are mutated in epidermolytic hyperkeratosis.Am J Hum Gen. 1994; 54: 179-190PubMed Google Scholar initially described a family in which the index patient was diagnosed with ichthyosis hystrix (localized) BCIE at the age of 5 mo and molecular analysis detected a L103Q mutation. Her son presented with blistering at birth and her granddaughter had blistering and generalized hyperkeratosis, especially over the flexural surfaces. Unfortunately, the phenotypic severity was not reported for this case and the histologic findings were not presented.Syder et al., 1994Syder A.J. Yu Q.C. Paller A.S. Giudice G. Pearson R. Fuchs E. Genetic mutations in the K1 and K10 genes of patients with epidermolytic hyperkeratosis.J Clin Invest. 1994; 93: 1533-1542Crossref PubMed Scopus (87) Google Scholar reported an atypically mild BCIE family that had a K100E mutation just proximal to the end of the 2B rod domain of K10. Although the clinical description did not report the development of annular and polycyclic erythematous, hyperkeratotic plaques, the histologic findings were somewhat similar to AEI. The third mutation in the 2B segment of K10 was discovered during our initial analysis of the molecular basis of AEI, and consisted of a R83E substitution (Joh et al., 1997Joh G.Y. Traupe H. Metze D. et al.A novel dinucleotide mutation in keratin 10 in the annular epidermolytic ichthyosis variant of bullous congenital ichthyosiform erythroderma.J Invest Dermatol. 1997; 108: 357-361Abstract Full Text PDF PubMed Scopus (43) Google Scholar). This mutation clearly resides outside of the putative overlap region of helix initiation and termination motifs. Therefore, it was not surprising that the clinical course of the disease in this family was much milder than typically described for BCIE patients. This study describes the fourth mutation, I107T, in the 2B segment of K10, which represents the first mutation in the helix termination motif of a type I keratin. Although the I107T mutation lies within the highly conserved helix termination motif, it does not appear to reside within the putative overlap region with the helix initiation motif. This isoleucine, however, is located in the "a" position of the heptad repeat and the substitution of a threonine would be expected to seriously impair the formation of a regular two-strand coiled-coil in this crucial part of the molecule. This may cause the molecule to terminate earlier than normal or the length may be the same as normal but the conformation may be less regular over the last 2–3 heptads. This could impede the next stage in the molecular hierarchy where the head-to-tail overlap occurs (Steinert and Parry, 1993Steinert P.M. Parry DaD The conserved H1 domain of the type II keratin I chain plays an essential role in the alignment of nearest neighbor molecules in mouse and human keratin1/keratin 10 intermediate filaments at the two-to-four-molecule level of structure.J Biol Chem. 1993; 268: 2878-2887Abstract Full Text PDF PubMed Google Scholar;Steinert et al., 1993Steinert P.M. Marekov L.N. Fraser Rdb Parry DaD Keratin intermediate filament structure: closslinking studies yield quantitative information on molecular dimensions and mechanism of assembly.J Mol Biol. 1993; 230: 436-452Crossref PubMed Scopus (230) Google Scholar). Having said this, it is surprising that the histopathologic findings in this case are so similar to IBS, i.e., cytolysis is restricted to the upper spinous and granular layers of the epidermis. This suggests that the dominant negative interfering effects of this K10 mutant are minimal in the lower spinous cells. Perhaps this particular K10 mutation does not substantially alter the keratin filament network until K5 and K14 subunits, which are known to persist into the spinous layer, are gradually depleted. Alternatively, this K10 mutant may have little effect on K1/K10 dimers, but primarily cause instability when it dimerizes with K2e, which is synthesized in the upper spinous and granular layers. Of course, this assumes that K2e has no other type I dimerization partner except K10. Clearly, there are still many unanswered questions concerning how individual keratin mutations cause disease. For this reason, it is important to continue to document the phenotypes caused by different keratin mutations. This information is not only useful for correlating genotypes with phenotypes in genetic counseling, but is also essential for inclusion in future molecular modeling and structure/function studies. We appreciate the interest and participation of the family members in this study. We also thank Drs. Joseph Rothnagel and David Parry for comments on the manuscript. This work was supported by NIH grant (HD25479) to D.R.R. Y.S. was partially supported by The Naito Foundation grant (94-453).
On reading the article by Mostow et al1in this issue of theArchives, it becomes apparent that several forces that will control our management of patients will converge on the issue of interferon for cutaneous T-cell lymphoma (CTCL).The New York Times(January 18, 1993:1) headline "New Standard for Drug Makers: Proving the Cure Is Worth the Cost" reflected Washington's interest in this and similar studies. Now, with an office in the White House, Hillary Clinton will take the lead in defining the integration of new therapies into the administration's health care policy. Prodding on this collision of interests is the revolution in T-cell biology that is bringing a hailstorm of biological insights into T-cell lymphoma faster than anyone can catch them or consider who should pay for them. These forces will encounter each other over the next few months, and therapy assessment will be a major issue.
