Operational classification of cutaneous squamous cell carcinomas based on unsupervised clustering of real cases by experts.

BACKGROUND: There is currently no staging system for cutaneous squamous cell carcinoma (cSCC) that is adapted to decision-making and universally used. Experts have unconscious ability to simplify the heterogeneity of clinical situations into a few relevant groups to drive their therapeutic decisions. Therefore, we have used unsupervised clustering of real cases by experts to generate an operational classification of cSCCs, an approach that was successful for basal cell carcinomas. OBJECTIVE: To generate a consensual and operational classification of cSCCs. METHOD: Unsupervised independent clustering of 248 cases of cSCCs considered difficult-to-treat. Eighteen international experts from different specialties classified these cases into what they considered homogeneous clusters useful for management, each with freedom regarding clustering criteria. Convergences and divergences between clustering were analysed using a similarity matrix, the K-mean approach and the average silhouette method. Mathematical modelling was used to look for the best consensual clustering. The operability of the derived classification was validated on 23 new practitioners. RESULTS: Despite the high heterogeneity of the clinical cases, a mathematical consensus was observed. It was best represented by a partition into five clusters, which appeared a posteriori to describe different clinical scenarios. Applicability of this classification was shown by a good concordance (94%) in the allocation of cases between the new practitioners and the 18 experts. An additional group of easy-to-treat cSCC was included, resulting in a six-group final classification: easy-to-treat/complex to treat due to tumour and/or patient characteristics/multiple/locally advanced/regional disease/visceral metastases. CONCLUSION: Given the methodology based on the convergence of unguided intuitive clustering of cases by experts, this new classification is relevant for clinical practice. It does not compete with staging systems, but they may complement each other, whether the objective is to select the best therapeutic approach in tumour boards or to design homogeneous groups for trials.

Position statement on classification of basal cell carcinomas. Part 2: EADO proposal for new operational staging system adapted to basal cell carcinomas.

BACKGROUND: No simple staging system has emerged for basal cell carcinomas (BCCs), since they do not follow the TNM process, and practitioners failed to agree on simple clinical or pathological criteria as a basis for a classification. Operational classification of BCCs is required for decision-making, trials and guidelines. Unsupervised clustering of real cases of difficult-to-treat BCCs (DTT-BCCs; part 1) has demonstrated that experts could blindly agree on a five groups classification of DTT-BCCs based on five patterns of clinical situations. OBJECTIVE: Using this five patterns to generate an operational and comprehensive classification of BCCs. METHOD: Testing practitioner's agreement, when using the five patterns classification to ensure that it is robust enough to be used in the practice. Generating the first version of a staging system of BCCs based on pattern recognition. RESULTS: Sixty-two physicians, including 48 practitioners and the 14 experts who participated in the generation of the five different patterns of DTT-BCCs, agreed on 90% of cases when classifying 199 DTT-BCCs cases using the five patterns classification (part 1) attesting that this classification is understandable and usable in practice. In order to cover the whole field of BCCs, these five groups of DTT-BCCs were added a group representing the huge number of easy-to-treat BCCs, for which sub-classification has little interest, and a group of very rare metastatic cases, resulting in a four-stage and seven-substage staging system of BCCs. CONCLUSION: A practical classification adapted to the specificities of BCCs is proposed. It is the first tumour classification based on pattern recognition of clinical situations, which proves to be consistent and usable. This EADO staging system version 1 will be improved step by step and tested as a decision tool and a prognostic instrument.

Position statement on classification of basal cell carcinomas. Part 1: unsupervised clustering of experts as a way to build an operational classification of advanced basal cell carcinoma based on pattern recognition.

BACKGROUND: No simple classification system has emerged for 'advanced basal cell carcinomas', and more generally for all difficult-to-treat BCCs (DTT-BCCs), due to the heterogeneity of situations, TNM inappropriateness to BCCs, and different approaches of different specialists. OBJECTIVE: To generate an operational classification, using the unconscious ability of experts to simplify the great heterogeneity of the clinical situations into a few relevant groups, which drive their treatment decisions. METHOD: Non-supervised independent and blinded clustering of real clinical cases of DTT-BCCs was used. Fourteen international experts from different specialties independently partitioned 199 patient cases considered 'difficult to treat' into as many clusters they want (≤10), choosing their own criteria for partitioning. Convergences and divergences between the individual partitions were analyzed using the similarity matrix, K-mean approach, and average silhouette method. RESULTS: There was a rather consensual clustering of cases, regardless of the specialty and nationality of the experts. Mathematical analysis showed that consensus between experts was best represented by a partition of DTT-BCCs into five clusters, easily recognized a posteriori as five clear-cut patterns of clinical situations. The concept of 'locally advanced' did not appear consistent between experts. CONCLUSION: Although convergence between experts was not granted, this experiment shows that clinicians dealing with BCCs all tend to work by a similar pattern recognition based on the overall analysis of the situation. This study thus provides the first consensual classification of DTT-BCCs. This experimental approach using mathematical analysis of independent and blinded clustering of cases by experts can probably be applied to many other situations in dermatology and oncology.

Pathological response and tumour bed histopathological features correlate with survival following neoadjuvant immunotherapy in stage III melanoma.

BACKGROUND: Guidelines for pathological evaluation of neoadjuvant specimens and pathological response categories have been developed by the International Neoadjuvant Melanoma Consortium (INMC). As part of the Optimal Neo-adjuvant Combination Scheme of Ipilimumab and Nivolumab (OpACIN-neo) clinical trial of neoadjuvant combination anti-programmed cell death protein 1/anti-cytotoxic T-lymphocyte-associated protein 4 immunotherapy for stage III melanoma, we sought to determine interobserver reproducibility of INMC histopathological assessment principles, identify specific tumour bed histopathological features of immunotherapeutic response that correlated with recurrence and relapse-free survival (RFS) and evaluate proposed INMC pathological response categories for predicting recurrence and RFS. PATIENTS AND METHODS: Clinicopathological characteristics of lymph node dissection specimens of 83 patients enrolled in the OpACIN-neo clinical trial were evaluated. Two methods of assessing histological features of immunotherapeutic response were evaluated: the previously described immune-related pathologic response (irPR) score and our novel immunotherapeutic response score (ITRS). For a subset of cases (n = 29), cellular composition of the tumour bed was analysed by flow cytometry. RESULTS: There was strong interobserver reproducibility in assessment of pathological response (κ = 0.879) and percentage residual viable melanoma (intraclass correlation coefficient = 0.965). The immunotherapeutic response subtype with high fibrosis had the strongest association with lack of recurrence (P = 0.008) and prolonged RFS (P = 0.019). Amongst patients with criteria for pathological non-response (pNR, >50% viable tumour), all who recurred had ≥70% viable melanoma. Higher ITRS and irPR scores correlated with lack of recurrence in the entire cohort (P = 0.002 and P ≤ 0.0001). The number of B lymphocytes was significantly increased in patients with a high fibrosis subtype of treatment response (P = 0.046). CONCLUSIONS: There is strong reproducibility for assessment of pathological response using INMC criteria. Immunotherapeutic response of fibrosis subtype correlated with improved RFS, and may represent a biomarker. Potential B-cell contribution to fibrosis development warrants further study. Reclassification of pNR to a threshold of ≥70% viable melanoma and incorporating additional criteria of <10% fibrosis subtype of response may identify those at highest risk of recurrence, but requires validation.

Prognostic impact of baseline tumour immune infiltrate on disease-free survival in patients with completely resected, BRAFv600 mutation-positive melanoma receiving adjuvant vemurafenib.

BACKGROUND: We conducted a retrospective exploratory analysis to evaluate the effects of baseline tumour immune infiltrate on disease-free survival (DFS) outcomes in patients with fully resected stage IIC-IIIC melanoma receiving adjuvant vemurafenib monotherapy or placebo in the BRIM8 study. PATIENTS AND METHODS: BRIM8 was a phase III, international, double-blind, randomised, placebo-controlled study. Eligible patients with BRAFV600 mutation-positive, completely resected melanoma were randomly assigned to oral vemurafenib (960 mg twice daily) or matching placebo for 52 weeks. The primary end point was DFS. The association of CD8+ T-cell infiltration and programmed death ligand 1 (PD-L1) expression with DFS, as measured by immunohistochemistry, was explored retrospectively. RESULTS: Four hundred ninety-eight patients were randomly assigned to receive adjuvant vemurafenib (n = 250) or placebo (n = 248); tumour samples were available for biomarker analysis for approximately 60% of patients. In the pooled biomarker population, placebo-treated patients with <1% CD8+ T cells in the tumour centre had shorter median DFS than those with ≥1% CD8+ T cells (7.7 versus 47.8 months). DFS benefit from vemurafenib versus placebo was greater in patients with <1% CD8+ T cells [hazard ratio (HR) 0.56; 95% confidence interval (CI) 0.34-0.92) than in patients with ≥1% CD8+ T cells (HR 0.77; 95% CI 0.48-1.22). Likewise, median DFS was shorter among placebo-treated patients with <5% versus ≥5% PD-L1+ immune cells (IC) in the tumour (7.2 versus 47.8 months). A greater DFS benefit with vemurafenib versus placebo was observed in patients with <5% PD-L1+IC (HR 0.36; 95% CI 0.24-0.56) than in patients with ≥5% PD-L1+IC (HR 0.99; 95% CI 0.58-1.69). CONCLUSIONS: The presence of CD8+ T cells and PD-L1+IC are favourable prognostic factors for DFS. Treatment with adjuvant vemurafenib may overcome the poor DFS prognosis associated with low CD8+ T-cell count or PD-L1 expression. CLINICALTRIALS. GOV IDENTIFIER: NCT01667419.

FDG-PET response and outcome from anti-PD-1 therapy in metastatic melanoma.

Background: Immune checkpoint inhibitor therapy has resulted in impressive and durable clinical activity for many cancers including melanoma; however, there remain few reliable predictors for long-term response. This study investigated whether [18F]2-fluoro-2-deoxy-D-glucose (FDG-PET) imaging may better predict long-term outcomes compared with standard computed tomography (CT) response criteria. Patients and methods: Retrospective analysis of metastatic melanoma patients treated with anti-PD-1-based immunotherapy with baseline and 1-year FDG-PET and CT imaging at Melanoma Institute Australia. One-year response was determined using RECIST for CT and EORTC criteria for PET, coded as complete response (CR or CMR), partial response (PR or PMR), stable disease (SD or SMD) or progressive disease (PD or PMD). Progression-free survival (PFS) was determined from the 1-year landmark. Results: Patients (n = 104) were evaluated with median follow-up 30.1 months and 98% remain alive. Most received anti-PD-1 as monotherapy (67%) or combined with ipilimumab (31%). At 1 year, 28% had CR, 66% had PR and 6% had SD on CT, while 75% had CMR, 16% PMR and 9% SMD/PMD on PET. CMR was observed in 68% of patients with PR on CT. RECIST PFS post 1-year landmark was similar in patients with CR versus PR/SD, but improved in patients with CMR versus non-CMR {median not reached [NR] versus 12.8 month; hazard ratio [HR] 0.06 [95% confidence interval (CI) 0.02-0.23]; P < 0.01}. In patients with PR on CT, PFS was improved in patients with PR + CMR versus PR + non-CMR (median NR versus 12.8 months; HR 0.07 [95% CI 0.02-0.27]; P < 0.01). In the 78 CMR patients, 78% had discontinued treatment and 96% had ongoing response. Conclusions: Whilst only a small proportion of patients have a CR at 1 year, most patients with a PR have CMR on PET. Almost all patients with CMR at 1 year have ongoing response to therapy thereafter. PET may have utility in predicting long-term benefit and help guide discontinuation of therapy.

Long-term efficacy and safety of sonidegib in patients with locally advanced and metastatic basal cell carcinoma: 30-month analysis of the randomized phase 2 BOLT study.

BACKGROUND: Patients with locally advanced basal cell carcinoma (laBCC) or metastatic BCC (mBCC), two difficult-to-treat populations, have had limited treatment options. Sonidegib, a hedgehog pathway inhibitor (HPI), was approved in laBCC based on results from the BOLT trial. OBJECTIVE: To evaluate long-term efficacy and safety of sonidegib in laBCC and mBCC in the BOLT 18- and 30-month analyses. METHODS: BOLT (NCT01327053, ClinicalTrials.gov), a double-blind phase 2 study, enrolled patients from July 2011 until January 2013. Eligible HPI-treatment-naïve patients with laBCC not amenable to curative surgery/radiotherapy or mBCC were randomized 1 : 2 to sonidegib 200 mg (laBCC, n = 66; mBCC, n = 13) or 800 mg (laBCC, n = 128; mBCC, n = 23). Tumour response was assessed per central and investigator review. RESULTS: With 30 months of follow-up, among patients treated with sonidegib 200 mg (approved dose), objective response rates were 56.1% (central) and 71.2% (investigator) in laBCC and 7.7% (central) and 23.1% (investigator) in mBCC. Tumour responses were durable as follows: median duration of response was 26.1 months (central) and 15.7 months (investigator) in laBCC and 24.0 months (central) and 18.1 months (investigator) in mBCC. Five patients with laBCC and three with mBCC in the 200-mg arm died. Median overall survival was not reached in either population; 2-year overall survival rates were 93.2% (laBCC) and 69.3% (mBCC). In laBCC, efficacy was similar regardless of aggressive or non-aggressive histology. Sonidegib 200 mg continued to have a better safety profile than 800 mg, with lower rates of grade 3/4 adverse events (43.0% vs. 64.0%) and adverse events leading to discontinuation (30.4% vs. 40.0%). CONCLUSION: Sonidegib continued to demonstrate long-term efficacy and safety in these populations. These data support the use of sonidegib 200 mg per local treatment guidelines.

Circulating tumour DNA predicts response to anti-PD1 antibodies in metastatic melanoma.

BACKGROUND: Programmed death 1 (PD1) inhibitors are now a foundation of medical management of metastatic melanoma. This study sought to determine whether circulating tumour DNA (ctDNA) provides useful early response and prognostic information. PATIENTS AND METHODS: We evaluated the relationship between pre-treatment and early on treatment ctDNA and outcome in melanoma patients treated with PD1 inhibitors alone or in combination with ipilimumab. RESULTS: ctDNA was detected in 40/76 patients (53%) at baseline, and correlated with stage, LDH levels, disease volume and ECOG performance. RECIST response was 72% (26/36) in group A (undetectable ctDNA at baseline), 77% (17/22) in group B (elevated ctDNA at baseline but undetectable within 12 weeks of therapy) and 6% (1/18) in group C (elevated ctDNA at baseline and remained elevated during treatment). The median PFS was not reached in groups A and B and was 2.7 months for group C [hazard ratio (HR) 0.09; P < 0.001 for group A versus C, and 0.16; P < 0.001 for group B versus C]. The median OS was not reached for groups A and B and was 9.2 months for group C (HR 0.02; P < 0.001 for group A versus C and 0.14; P < 0.001 for group B versus C). The poor outcome measures associated with group C remained significant in multivariate analysis adjusted for LDH, performance status, tumour stage and disease volume. The predictive value for ctDNA for response was confirmed in a separate validation cohort (n = 29, P < 0.01). CONCLUSION: Longitudinal assessment of ctDNA in metastatic melanoma patients receiving treatment with PD1 inhibitors is an accurate predictor of tumour response, PFS and OS. Patients who had a persistently elevated ctDNA on therapy had a poor prognosis, and this may guide combination and sequencing of subsequent therapies.

Anti-PD-1 therapy in patients with advanced melanoma and preexisting autoimmune disorders or major toxicity with ipilimumab.

Background: Anti-PD-1 antibodies (anti-PD-1) have clinical activity in a number of malignancies. All clinical trials have excluded patients with significant preexisting autoimmune disorders (ADs) and only one has included patients with immune-related adverse events (irAEs) with ipilimumab. We sought to explore the safety and efficacy of anti-PD-1 in such patients. Patients and methods: Patients with advanced melanoma and preexisting ADs and/or major immune-related adverse events (irAEs) with ipilimumab (requiring systemic immunosuppression) that were treated with anti-PD-1 between 1 July 2012 and 30 September 2015 were retrospectively identified. Results: One hundred and nineteen patients from 13 academic tertiary referral centers were treated with anti-PD-1. In patients with preexisting AD (N = 52), the response rate was 33%. 20 (38%) patients had a flare of AD requiring immunosuppression, including 7/13 with rheumatoid arthritis, 3/3 with polymyalgia rheumatica, 2/2 with Sjogren's syndrome, 2/2 with immune thrombocytopaenic purpura and 3/8 with psoriasis. No patients with gastrointestinal (N = 6) or neurological disorders (N = 5) flared. Only 2 (4%) patients discontinued treatment due to flare, but 15 (29%) developed other irAEs and 4 (8%) discontinued treatment. In patients with prior ipilimumab irAEs requiring immunosuppression (N = 67) the response rate was 40%. Two (3%) patients had a recurrence of the same ipilimumab irAEs, but 23 (34%) developed new irAEs (14, 21% grade 3-4) and 8 (12%) discontinued treatment. There were no treatment-related deaths. Conclusions: In melanoma patients with preexisting ADs or major irAEs with ipilimumab, anti-PD-1 induced relatively frequent immune toxicities, but these were often mild, easily managed and did not necessitate discontinuation of therapy, and a significant proportion of patients achieved clinical responses. The results support that anti-PD-1 can be administered safely and can achieve clinical benefit in patients with preexisting ADs or prior major irAEs with ipilimumab.

Activity and safety of radiotherapy with anti-PD-1 drug therapy in patients with metastatic melanoma.

The anti-PD-1 antibodies nivolumab and pembrolizumab are active in metastatic melanoma; however, there is limited data on combining anti-PD-1 antibody and radiotherapy (RT). We sought to review clinical outcomes of patients receiving RT and anti-PD-1 therapy. All patients receiving anti-PD-1 antibody and RT for metastatic melanoma were identified. RT and systemic treatment, clinical outcome, and toxicity data were collected. Fifty-three patients were included; 35 patients received extracranial RT and/or intracranial stereotactic radiosurgery (SRS) and 21 received whole brain radiotherapy (WBRT) (three of whom also received SRS/extracranial RT). Patients treated with extracranial RT or SRS received treatment either sequentially (RT then anti-PD-1, n = 11), concurrently (n = 16), or concurrent "salvage" treatment to lesions progressing on anti-PD-1 therapy (n = 15). There was no excessive anti-PD-1 or RT toxicity observed in patients receiving extracranial RT. Of six patients receiving SRS, one patient developed grade 3 radiation necrosis. In 21 patients receiving WBRT, one patient developed Stevens-Johnson syndrome, one patient developed acute neurocognitive decline, and one patient developed significant cerebral edema in the setting of disease. Response in irradiated extracranial/intracranial SRS lesions was 44% for sequential treatment and 64% for concurrent treatment (p=0.448). Likewise there was no significant difference between sequential or concurrent treatment in lesional response of non-irradiated lesions. For progressing lesions subsequently irradiated, response rate was 45%. RT and anti-PD-1 antibodies can be safely combined, with no detectable excess toxicity in extracranial sites. WBRT and anti-PD-1 therapy is well tolerated, although there are rare toxicities and the role of either anti-PD-1 or WBRT in the etiology of these is uncertain.

Efficacy and toxicity of treatment with the anti-CTLA-4 antibody ipilimumab in patients with metastatic melanoma after prior anti-PD-1 therapy.

BACKGROUND: Recent phase III clinical trials have established the superiority of the anti-PD-1 antibodies pembrolizumab and nivolumab over the anti-CTLA-4 antibody ipilimumab in the first-line treatment of patients with advanced melanoma. Ipilimumab will be considered for second-line treatment after the failure of anti-PD-1 therapy. METHODS: We retrospectively identified a cohort of 40 patients with metastatic melanoma who received single-agent anti-PD-1 therapy with pembrolizumab or nivolumab and were treated on progression with ipilimumab at a dose of 3 mg kg(-1) for a maximum of four doses. RESULTS: Ten percent of patients achieved an objective response to ipilimumab, and an additional 8% experienced prolonged (>6 months) stable disease. Thirty-five percent of patients developed grade 3-5 immune-related toxicity associated with ipilimumab therapy. The most common high-grade immune-related toxicity was diarrhoea. Three patients (7%) developed grade 3-5 pneumonitis leading to death in one patient. CONCLUSIONS: Ipilimumab therapy can induce responses in patients who fail the anti-PD-1 therapy with response rates comparable to previous reports. There appears to be an increased frequency of high-grade immune-related adverse events including pneumonitis that warrants close surveillance.

Phase II study of single-agent panitumumab in patients with incurable cutaneous squamous cell carcinoma.

BACKGROUND: Although advanced cutaneous squamous cell carcinoma (CSCC) is quite common, there are few prospective trials regarding its optimal management. This study evaluated the efficacy and safety of single-agent panitumumab in the treatment of patients with CSCC not suitable for local therapy. PATIENTS AND METHODS: Sixteen patients received single-agent panitumumab at a dose of 6 mg/kg repeated every 2 weeks for a minimum of three cycles and continued until progression, a maximum of nine cycles or dose-limiting toxicity. The primary end point was the best overall response rate (ORR) as assessed by Response Evaluation Criteria in Solid Tumours (RECIST version 1.1) criteria. Secondary end points included evaluation of safety, toxicity and progression-free survival (PFS). RESULTS: Between May 2010 and May 2012, 16 patients were recruited. Fourteen patients were male and the median age was 68 years. Fifteen patients had locoregionally advanced or recurrent disease with 14 patients receiving previous radiotherapy and 7 receiving previous cytotoxic chemotherapy. The best ORR [partial (PR) or complete response (CR)] was 31% (3/16 PR, 2/16 CR) with a further 6 of 16 patients achieving SD. The median PFS and overall survival were 8 and 11 months respectively. Grade 3 or 4 events were observed in five patients (four being skin toxicity) with one patient ceasing due to skin toxicity. With a median follow-up of 24 months, 10 patients died due to progressive disease, 6 are alive, one patient with no evidence of disease at the time of analysis. CONCLUSIONS: Single-agent panitumumab is safe and effective in the management of patients with advanced CSCC even in a previously extensively pre-treated cohort.

Preclinical evaluation of dual PI3K-mTOR inhibitors and histone deacetylase inhibitors in head and neck squamous cell carcinoma.

BACKGROUND: We examine the potential value of a series of clinically relevant PI3K-mTOR inhibitors alone, or in combination with histone deacetylase inhibitors, in a model of head and neck squamous cell carcinoma (HNSCC). METHODS: Head and neck squamous cell carcinoma cell lines, human keratinocyte and HNSCC xenograft models were treated with histone deacetylase inhibitors (HDACIs) and new generation PI3K and dual PI3K-mTOR inhibitors either alone or in combination. Cell and tumour tissue viability and proliferation were then determined in vitro and in vivo. RESULTS: Phosphatidylinositol-3-phosphate kinase, AKT and dual PI3K-mTOR inhibitors caused marked in vitro enhancement of cytotoxicity induced by HDACIs in HNSCC cancer cells. This effect correlates with AKT inhibition and is attenuated by expression of constitutively active AKT. Histone deacetylase inhibitor and phosphatidylinositol-3-phosphate kinase inhibitors (PI3KIs) inhibited tumour growth in xenograft models of HNSCC. Importantly, we observed intratumoural HDAC inhibition and PI3K inhibition as assessed by histone H3 acetylation status and phospho-AKT staining, respectively. However, we saw no evidence of improved efficacy with an HDACI/PI3KI combination. INTERPRETATION: That PI3K and dual PI3K-mTOR inhibitors possess antitumour effect against HNSCC in vivo.

Indole-3-carbinol - induced growth inhibition can be converted to a cytotoxic response in the presence of TPA+Ca(2+) in squamous cell carcinoma cell lines.

We examined the possibility that I3C, when combined with a differentiation stimulus (TPA+CaCl(2)), would sensitise SCC cells to a differentiation stimulus. We report that I3C induces a profound growth inhibition in SCC cells that is dissimilar to the growth inhibition required to initiate differentiation. Moreover, we report that I3C, when combined with TPA+CaCl(2) treatment, induces a loss of colony forming ability that was differentiation and senescence - independent but was due to delayed cytotoxicity. This study shows that I3C in combination with a PKC activator+Ca(2+) may be a useful therapeutic strategy for treating oral SCC.

Carboplatin and paclitaxel interact antagonistically in a megakaryoblast cell line--a potential mechanism for paclitaxel-mediated sparing of carboplatin-induced thrombocytopenia.

PURPOSE: Clinical observation has shown that paclitaxel ameliorates the antiplatelet toxicity of carboplatin when the two drugs are combined, although antitumour activity and white cell toxicity are at least additive. We hypothesized that this is due to an interaction between the two drugs at the level of the platelet precursor. METHODS: We measured inhibition of growth of the megakaryoblast cell line MEG-01 following exposure to paclitaxel and carboplatin singly or combined. Drug interaction was assessed by median effect analysis. RESULTS: An antagonistic interaction was observed, and this was most marked at drug concentrations giving a low level of growth inhibition (P < 0.002, sign test). The interaction was not sequence-dependent. There was no significant difference in whole-cell accumulation of platinum or the amount of platinum adducts on DNA following combined treatment in comparison with carboplatin alone. CONCLUSIONS: These results provide the first evidence of an antagonistic interaction between paclitaxel and carboplatin in a platelet precursor and provide an explanation for the platelet-sparing effect of the combination of these chemotherapeutic agents. While the mechanisms underlying the interaction described in this report are yet to be fully elucidated, this study provides evidence that the antagonism between paclitaxel and carboplatin in MEG-01 cells is not due to reduced platination of DNA.