Portal de Periódicos da CAPES

The Usefulness of Teledermatology in Monitoring the Efficacy and Safety of a Topical Treatment for Actinic Keratosis

2021; Elsevier BV; Volume: 141; Issue: 7 Linguagem: Inglês

10.1016/j.jid.2020.12.020

1523-1747

J. Sola-Ortigosa, C. Muñoz‐Santos, Isabel Otzet-Gramunt, Teresa Masat-Ticó, Joan Isidro-Ortega, Antonio Guilabert, TMT, JIO, Eustaqui Hernandez-Simón, MaTeresa Rodellar-Oncins, Maite Arteaga-Salinas, IOG, Enric HernÁndez-Huet, Mireia RosÁs-Redondo, Ana Maria Pérez-López, Carolina Echegaray-Carranza, Jaume Montal-Codina, Montserrat Argila-Vicente, Jordi Sobrequés-Soriano, Ariel León-Pérez, Pilar Perramon-Rovira, Mireia Prat-Benítez, Amada Aguyé, Araceli Bautista-Puente, Ricardo SÁnchez-Pe ́rez, Francisco Leiva-Morote, Gisela Solsona-TomÁs, Josep Oliveras-Cuquet, Carlota Arroyo-Rodríguez,

Cutaneous Melanoma Detection and Management

We recently reported that teledermatology (TD) is valid and reliable in the diagnosis of actinic keratosis (AK) and field cancerization (Sola-Ortigosa et al., 2020Sola-Ortigosa J. Muñoz-Santos C. Masat-Ticó T. Isidro-Ortega J. Guilabert A. Grup d'Estudi de Teledermatologia del Vallès OrientalThe role of teledermatology and teledermoscopy in the diagnosis of actinic keratosis and field cancerization.J Invest Dermatol. 2020; 140: 1976-1984.e4Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar). Although AK is mostly treated by dermatologists, primary care physicians (PCPs) participate in the management and follow-up (de Berker et al., 2017de Berker D. McGregor J.M. Mohd Mustapa M.F. Exton L.S. Hughes B.R. British Association of Dermatologists' guidelines for the care of patients with actinic keratosis 2017.Br J Dermatol. 2017; 176: 20-43Crossref PubMed Scopus (100) Google Scholar; Lowell et al., 2001Lowell B.A. Froelich C.W. Federman D.G. Kirsner R.S. Dermatology in primary care: prevalence and patient disposition.J Am Acad Dermatol. 2001; 45: 250-255Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). Topical AK treatments usually produce local skin reactions (LSRs) that lead to unscheduled dermatological consultations and treatment withdrawal. We hypothesized that TD could be used for monitoring topical AK therapy in the primary care setting, thus reducing treatment-related referrals. The objectives of this study were to test TD as a tool for AK count, analyze the validity and reliability of TD in the clinical response to an AK topical treatment (ingenol mebutate [IMB]), and study by TD the LSR pattern of IMB in a real-life scenario of a gatekeeper healthcare system (Supplementary Materials and Methods). To test TD for AK count, 6,249 AK within 260 cancerization fields from 242 patients were evaluated (Figure 1a). Overall, TD showed high performance for AK count. Intraclass correlation coefficients were greater than 0.88, and Bland‒Altman plots verified that the mean bias in the AK count in all cases was within the tolerance interval (mean ± 1.96[SD]). Diagnostic test performance parameters and intraobserver and/or interobserver agreement were also high for AK count (Supplementary Table S1 and Supplementary Figure S1). A total of 65 patients with 88 cancerization fields received IMB (Figure 1b). Treated patient's characteristics are shown in Supplementary Table S2. TD was valid in the clinical response evaluation (Table 1). Complete and partial clearance AK reduction and mean AK count reduction after treatment agreements were good or excellent, with the intraclass correlation coefficients being greater than 0.94, 0.91, and 0.87, respectively. Bland‒Altman plots showed good agreement (Supplementary Figure S2). Diagnostic test performance parameters produced similar results. The interobserver agreement also achieved good results among the three teledermatologists (intraclass correlation coefficient > 0.88) (Supplementary Table S3 and Supplementary Figure S3). See example pictures of AK count using TD in Supplementary Figure S4.Table 1Clinical Response Agreement and Diagnostic Test Performance Parameters of TD in 88 Cancerization Fields from 65 PatientsClinical Response Agreement of TDClinical Response ParametersICC (95% CI)P-ValueCCR1CCR: 100% of reduction of AK. (n = 31 fields) TDA0.97 (0.95–0.99)<0.001 TDB0.95 (0.92–0.97)<0.001 TDC0.94 (0.91–0.96)<0.001PCR2PCR: ≥75% of reduction of AK. (n = 57 fields) TDA0.92 (0.89–0.94)<0.001 TDB0.93 (0.90–0.95)<0.001 TDC0.91 (0.88–0.93)<0.001Mean reduction in AK count (n = 88 fields) TDA0.87 (0.81–0.92)<0.001 TDB0.92 (0.88–0.95)<0.001 TDC0.91 (0.87–0.94)<0.001Diagnostic Test Performance Parameters of TD for Clinical ResponseClinical Response ParametersSensitivity,% (95% CI)Specificity,% (95% CI)PPV,% (95% CI)NPV,% (95% CI)CCR1CCR: 100% of reduction of AK. TDA96.7 (95.1–98.2)96.5 (94.0–97.8)96.7 (94.9–98.0)97.8 (96.0–99.1) TDB97.5 (95.9–98.0)96.7 (94.9–98.0)96.7 (95.0–97.8)97.6 (95.8–98.9) TDC97.1 (95.3–98.4)97.4 (95.6–98.7)97.7 (95.9–99.0)97.4 (96.6–98.7) Mean value97.1 (95.4–98.2)96.9 (94.9–98.1)97.0 (95.2–98.3)97.6 (96.1–98.9)PCR2PCR: ≥75% of reduction of AK. TDA86.4 (84.6–87.7)87.4 (85.6–88.7)87.2 (85.4–88.5)87.1 (85.3–88.4) TDB89.4 (87.7–90.8)87.1 (85.3–88.4)87.2 (85.4–88.5)88.6 (86.8–89.9) TDC88.8 (86.3–89.9)86.7 (84.9–88.0)87.4 (85.6–88.7)87.9 (86.1–89.2) Mean value88.2 (86.2–89.5)87.1 (85.3–88.4)87.3 (85.5–88.6)87.9 (86.1–89.2) P-value TD-CCR1CCR: 100% of reduction of AK. vs PCR2PCR: ≥75% of reduction of AK. 20 mm to <30 mm were counted as three AK. Intraobserver agreement on the baseline (pretreatment) AK count was defined as agreement by the same teledermatologist between stages 1 and 2. Interobserver agreement was defined as an agreement between the three teledermatologists using a mean of stages 1 and 2 (Figure 1a). In stage 3, the three dermatologists clinically evaluated the baseline AK count separately and then jointly to reach a consensus. TD agreement for the baseline AK count was calculated using the agreed FTF consensus as the pragmatic reference standard. Diagnostic test performance parameters were calculated. Post-FTF consultation, patients diagnosed with isolated AK or field cancerization were offered all available therapeutic options for AK at our hospital (Figure 1b). Assignment to IMB treatment was not decided in advance but determined by clinical practice and the patient's preference. We considered IMB the best candidate for the study because it is an effective short-course, field-directed therapy with easy dosing and has a favorable safety profile with predictable local skin reactions (LSRs) and good adherence, which permits a more predictable and homogeneous TD-based follow-up (Hofbauer et al., 2014Hofbauer G. Anliker M. Boehncke W.H. Brand C. Braun R. Gaide O. et al.Swiss clinical practice guidelines on field cancerization of the skin.Swiss Med Wkly. 2014; 144: w14026PubMed Google Scholar; Lebwohl et al., 2013Lebwohl M. Shumack S. Stein Gold L. Melgaard A. Larsson T. Tyring S.K. Long-term follow-up study of ingenol mebutate gel for the treatment of actinic keratoses.JAMA Dermatol. 2013; 149: 666-670Crossref PubMed Scopus (122) Google Scholar; Stockfleth et al., 2015Stockfleth E. Peris K. Guillen C. Cerio R. Basset-seguin N. Foley P. et al.Review: a consensus approach to improving patient adherence and persistence with topical treatment for actinic keratosis.Int J Dermatol. 2015; 54: 509-515Crossref PubMed Scopus (30) Google Scholar; Werner et al., 2015Werner R.N. Jacobs A. Rosumeck S. Erdmann R. Sporbeck B. Methods and results report - evidence and consensus- based (S3) guidelines for the treatment of actinic keratosis -International League of Dermatological Societies in cooperation with the European Dermatology Forum.J Eur Acad Dermatol Venereol. 2015; 29: e1-e66Crossref Scopus (25) Google Scholar). Patients were informed of the dosing regimens and LSR of each therapy. IMB was chosen because of its very easy and comfortable regimen of 2–3 days and the predictable, well-established LSR in pivotal studies (Lebwohl et al., 2013Lebwohl M. Shumack S. Stein Gold L. Melgaard A. Larsson T. Tyring S.K. Long-term follow-up study of ingenol mebutate gel for the treatment of actinic keratoses.JAMA Dermatol. 2013; 149: 666-670Crossref PubMed Scopus (122) Google Scholar, Lebwohl et al., 2012Lebwohl M. Swanson N. Anderson L.L. Melgaard A. Xu Z. Berman B. Ingenol mebutate gel for actinic keratosis.N Engl J Med. 2012; 366: 1010-1019Crossref PubMed Scopus (388) Google Scholar). Only patients with field cancerization were offered IMB therapy. Patients choosing IMB who agreed to participate gave written informed consent and were included in the treatment phase (stage 4). The study complied with the Declaration of Helsinki and was approved by the Hospital General de Granollers, Barcelona, Spain and Primary Care Ethics Committees. Patients consented to the publication of images of TD. IMB was administered following clinical practice. One tube was prescribed for each 25 cm2 area: 500 μg/g or 150 μg/g presentations were applied to the trunk or extremities for 2 consecutive days or to the face or scalp for 3 days, respectively. More than one field cancerization could be treated simultaneously. Exclusion criteria for IMB treatment were as follows: (i) AK in high-risk locations with a high probability of invasive squamous cell carcinoma requiring confirmatory biopsy; (ii) open wounds or damaged skin, near the eyes or the nose, inside the ears or lips; (iii) AK grade 3 (hypertrophic, very noticeable to touch, and visible); and (iv) Other contraindications according to the IMB datasheet. Before study initiation, PCP received specific training on IMB therapy and the expected LSR. Patients were followed and controlled by PCP after IMB prescription in the dermatologist FTF consultation. IMB was applied the first day on Sunday afternoon, and PCP took clinical photographs of LSR from Monday to Friday and on days 8, 15, and 57 (Figure 1b). PCP received a report every 3 days with recommendations for LRS care (e.g., the use of petrolatum and avoidance of topical corticosteroids). All images were evaluated separately by the three teledermatologists. On day 60, patients attended a hospital FTF consultation, which was evaluated by the three dermatologists (stage 5, Figure 1b). First, each dermatologist recorded the final (post-treatment) AK count separately. Consensus on the final number of AK was reached and considered the pragmatic reference standard (agreed FTF evaluation). Agreement on the clinical response of TD was defined as the agreement between TD at day 57 and the FTF consultation at day 60 with respect to the AK reduction after IMB treatment. AK count reduction was calculated comparing the recorded agreed FTF evaluation for the baseline AK count (stage 3) with the final count either by TD or FTF consultation. Complete clearance was considered when no clinically visible AK lesions within the treatment area with respect to baseline occurred, and partial clearance was defined as ≥75% of clinical AK cleared within the treatment area with respect to baseline. Mean reduction in the AK count was also evaluated. Diagnostic test performance parameters were calculated for TD with respect to these outcomes. Interobserver agreement of TD for clinical response was calculated. We analyzed the capacity of TD to monitor IMB LSR in primary care. LSRs were evaluated separately by the three teledermatologists. LSR timing and duration and the six categories of signs (erythema, flaking and/or scaling, crusting, swelling, vesiculation and/or postulation, and erosion and/or ulceration) were evaluated, with sign graded from 0 (absent) to 4 (maximum response). A composite-validated LSR score representing the sum of the six individual scores was calculated, giving a score of 0–24 at each assessment (Rosen et al., 2014Rosen R. Marmur E. Anderson L. Welburn P. Katsamas J. A new, objective, quantitative scale for measuring local skin responses following topical actinic keratosis therapy with ingenol mebutate.Dermatol Ther (Heidelb). 2014; 4: 207-219Crossref PubMed Scopus (55) Google Scholar). Interobserver agreement of TD for LSR was calculated using the mean LSR score value of each teledermatologist in day 3–8 visits. Quantitative data were represented using descriptive statistics. All agreements were assessed using intraclass correlation coefficients, with values of 0.9 indicative of poor, moderate, good, and excellent or perfect agreement, respectively (Gisev et al., 2013Gisev N. Bell J.S. Chen T.F. Interrater agreement and interrater reliability: key concepts, approaches, and applications.Res Soc Adm Pharm. 2013; 9: 330-338Crossref PubMed Scopus (294) Google Scholar), and Bland‒Altman plots to describe agreements between the two quantitative measurements by constructing the limits of agreement (mean difference of bias ± 1.96 [SD] of the differences between the two measurements) (Bland and Altman, 1986Bland J.M. Altman D.G. Statistical methods for assessing agreement between two methods of clinical measurement.Lancet. 1986; 1: 307-310Abstract PubMed Scopus (38558) Google Scholar). Furthermore, good agreement between the two measures would be indicated by a mean difference close to zero without obvious trends in the plotted points between the difference and average of these two assessments. Sensitivity, specificity, and positive and negative predictive values for TD were calculated for the AK count and clinical response. All significance tests were two tailed with P < 0.05 considered statistically significant. Statistical analysis was made using the Statistical Package for the Social Sciences statistics, version 24.0 (IBM, Armonk, NY) and R, version 3.4 (R Core Team, Vienna, Austria).Supplementary Figure S2Bland‒Altman plots of clinical response agreement of TD by the three teledermatologists. AK, actinic keratosis; CCR, complete clearance response; PCR, partial clearance response; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S3Bland‒Altman plots of inter. agreement of TD for clinical response by the three teledermatologists. AK, actinic keratosis; CCR, complete clearance response; inter., interobserver; PCR, partial clearance response; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S4Example pictures of AK count using TD. (a) Baseline field cancerization; (b) Baseline AK count; (c) Final field cancerization; (d) Final AK count. The patient consented to the publication of the images. AK, actinic keratosis; TD, teledermatology.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S5Bland‒Altman plots of inter. agreement of TD for LSRs evaluation (LSR score) by the three teledermatologists. LSR, local skin reaction; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S6Time course of LSR. (a) Time course of mean composite LSR scores in 65 patients treated with IMB with 88 cancerization fields by the three teledermatologists in the face and/or scalp and trunk and/or extremities. Composite scores represent the sum of six individual LSR scores calculated at each study visit for each teledermatologist. (b) Time course of mean composite LSR scores from three clinical trials (Anderson et al., 2009Anderson L. Schmieder G.J. Werschler W.P. Tschen E.H. Ling M.R. Stough D.B. et al.Randomized, double-blind, double-dummy, vehicle-controlled study of ingenol mebutate gel 0.025% and 0.05% for actinic keratosis.J Am Acad Dermatol. 2009; 60: 934-943Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar; Lebwohl et al., 2012Lebwohl M. Swanson N. Anderson L.L. Melgaard A. Xu Z. Berman B. Ingenol mebutate gel for actinic keratosis.N Engl J Med. 2012; 366: 1010-1019Crossref PubMed Scopus (388) Google Scholar; Rosen et al., 2014Rosen R. Marmur E. Anderson L. Welburn P. Katsamas J. A new, objective, quantitative scale for measuring local skin responses following topical actinic keratosis therapy with ingenol mebutate.Dermatol Ther (Heidelb). 2014; 4: 207-219Crossref PubMed Scopus (55) Google Scholar). IMB, ingenol mebutate; LSR, local skin reaction; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S7Individual LSR signs over time by TD using a mean of the three teledermatologists. Each LSR sign is graded from 0 (not present) to 4 (maximum possible response). LSR, local skin reaction; TD, teledermatology.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S8Example pictures of LSR provided by the PCP using TD. (a) day 1; (b) day 3; (c) day 5; (d) day 8; (e) day 15; (f) day 57. The patient consented to the publication of the images. LSR, local skin reaction; PCP, primary care physician; TD, teledermatology.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S9Standard form of clinical and demographic characteristics filled by the PCPs. PCP, primary care physician; TALENEO, Teledermatology cutaneous Neoplasms Study.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Table S1Agreement, Diagnostic Tests Performance Parameters, and Intraobserver and Interobserver Agreements for Baseline AK Count by TDA, TDB, and TDC in 260 Cancerization Fields from 242 PatientsAgreement of TD for AK CountTeledermatologistICC (95% CI)P-ValueTDA0.92 (0.89–0.94)<0.001TDB0.88 (0.84–0.93)<0.001TDC0.89 (0.86–0.92)<0.001Diagnostic Test Performance Parameters of TD for AK CountTeledermatologistSensitivity,% (95% CI)Specificity,% (95% CI)PPV,% (95% CI)NPV,% (95% CI)TDA83.2 (81.4–86.5)88.8 (86.3–89.9)88.6 (86.8–89.9)87.8 (85.0–89.1)TDB84.5 (82.6–87.1)87.9 (86.1–89.2)86.7 (84.9–88.2)85.6 (82.8–88.9)TDC85.1 (83.3–87.4)89.4 (87.7–90.8)84.7 (82.9–86.0)87.4 (84.6–89.7)Intraobserver Agreement of TD for AK CountTeledermatologistICC (95% CI)P-ValueTDA0.94 (0.92–0.95)<0.001TDB0.95 (0.93–0.96)<0.001TDC0.91 (0.89–0.92)<0.001Interobserver Agreement of TD for AK CountTeledermatologistICC (95% CI)P-ValueTDA versus TDB0.93 (0.92–0.94)<0.001TDA versus TDC0.94 (0.92–0.95)<0.001TDB versus TDC0.91 (0.90–0.92)<0.001Abbreviations: AK, actinic keratosis; CI, confidence interval; ICC, intraclass correlation coefficient; NPV, negative predictive value; PPV, positive predictive value; TD, teledermatology; TDA, Teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C. Open table in a new tab Supplementary Table S2Characteristics of Study Participants and Cancerization Fields in the Treatment Phase with IMBCharacteristicsN = 65 PatientsMean age, y, mean (SD)75.4 (7.4)Gender, n (%) Male50 (76.9) Female15 (23.1)Ethnicity, n (%), Caucasian65 (100)Fitzpatrick skin type I11 (16.9) II29 (44.6) III21 (32.3) IV4 (6.2) V0Sun exposure due to profession, n (%) Construction workers13 (20) Farm workers14 (21.5) Sea workers0 Recreational beach activities10 (15.4) Mountain recreational activities3 (4.6) No history of sun exposure25 (38.5)Baldness, n (%) Total12 (18.5) Partial31 (47.7) Without baldness22 (33.8)N= 88 Cancerization FieldsLocation, n (%) Face38 (43.2)Right face7 (8.0)Left face18 (20.4)Forehead10 (11.4)Nose3 (3.4) Scalp48 (54.5) Neckline0 Hands and arms0 Pretibial areas limbs2 (2.3)Abbreviation: IMB, ingenol mebutate. Open table in a new tab Supplementary Table S3Interobserver Agreement of TD for Clinical Response in 88 Cancerization Fields from 65 PatientsClinical Response ParametersICC (95% CI)P-ValueComplete clearance (n = 31 fields) TDA versus TDB0.94 (0.92–0.95)<0.001 TDA versus TDC0.93 (0.92–0.94)<0.001 TDB versus TDC0.88 (0.83–0.92)<0.001Partial clearance (n = 57 fields) TDA versus TDB0.93 (0.91–0.94)<0.001 TDA versus TDC0.91 (0.89–0.92)<0.001 TDB versus TDC0.89 (0.84–0.93)<0.001Mean reduction in AK count (n = 88 fields) TDA versus TDB0.91 (0.87–0.94)<0.001 TDA versus TDC0.95 (0.93–0.97)<0.001 TDB versus TDC0.89 (0.84–0.93)<0.001Abbreviations: AK, actinic keratosis; CI, confidence interval; ICC, intraclass correlation coefficient; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C. Open table in a new tab Supplementary Table S4Interobserver Agreement of TD for LSRs Evaluation by LSR Score in 88 Cancerization Fields from 65 PatientsInterobserver AgreementICC95% CIP-ValueTDA versus TDB0.89 (0.84–0.92)<0.001TDA versus TDC0.90 (0.85–0.93)<0.001TDB versus TDC0.92 (0.88–0.94)<0.001Abbreviations: CI, confidence interval; ICC, intraclass correlation coefficient; LSR, local skin reaction; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C. Open table in a new tab Abbreviations: AK, actinic keratosis; CI, confidence interval; ICC, intraclass correlation coefficient; NPV, negative predictive value; PPV, positive predictive value; TD, teledermatology; TDA, Teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C. Abbreviation: IMB, ingenol mebutate. Abbreviations: AK, actinic keratosis; CI, confidence interval; ICC, intraclass correlation coefficient; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C. Abbreviations: CI, confidence interval; ICC, intraclass correlation coefficient; LSR, local skin reaction; TD, teledermatology; TDA, teledermatologist A; TDB, teledermatologist B; TDC, teledermatologist C.