Validating whole slide imaging for diagnostic purposes in pathology: guideline from the College of American Pathologists Pathology and Laboratory Quality Center

CONTEXT

There is increasing interest in using whole slide imaging (WSI) for diagnostic purposes (primary and/or consultation). An important consideration is whether WSI can safely replace conventional light microscopy as the method by which pathologists review histologic sections, cytology slides, and/or hematology slides to render diagnoses. Validation of WSI is crucial to ensure that diagnostic performance based on digitized slides is at least equivalent to that of glass slides and light microscopy. Currently, there are no standard guidelines regarding validation of WSI for diagnostic use.

OBJECTIVE

To recommend validation requirements for WSI systems to be used for diagnostic purposes.

DESIGN

The College of American Pathologists Pathology and Laboratory Quality Center convened a nonvendor panel from North America with expertise in digital pathology to develop these validation recommendations. A literature review was performed in which 767 international publications that met search term requirements were identified. Studies outside the scope of this effort and those related solely to technical elements, education, and image analysis were excluded. A total of 27 publications were graded and underwent data extraction for evidence evaluation. Recommendations were derived from the strength of evidence determined from 23 of these published studies, open comment feedback, and expert panel consensus.

RESULTS

Twelve guideline statements were established to help pathology laboratories validate their own WSI systems intended for clinical use. Validation of the entire WSI system, involving pathologists trained to use the system, should be performed in a manner that emulates the laboratory's actual clinical environment. It is recommended that such a validation study include at least 60 routine cases per application, comparing intraobserver diagnostic concordance between digitized and glass slides viewed at least 2 weeks apart. It is important that the validation process confirm that all material present on a glass slide to be scanned is included in the digital image.

CONCLUSIONS

Validation should demonstrate that the WSI system under review produces acceptable digital slides for diagnostic interpretation. The intention of validating WSI systems is to permit the clinical use of this technology in a manner that does not compromise patient care.

In vitro and in vivo comparison of two suprachoroidal shunts

PURPOSE

To compare fibrosis, aqueous humor dynamics, and intraocular pressure (IOP) of two suprachoroidal shunts as part of a new class of glaucoma drainage devices.

METHODS

Following proliferation analysis in vitro, 20 rabbits were implanted with either a gold shunt (GS, GMSplus+, SOLX) or a polypropylene shunt (PS, Aquashunt, OPKO). Ten eyes received mitomycin C (MMC) and triamcinolone. Peak and trough IOP were monitored with a pneumatonometer and tono-pen for 15 weeks. Aqueous humor dynamics were evaluated fluorophotometrically and tonographically. Fibrosis was quantified.

RESULTS

In vitro proliferation was similar. In vivo, both shunts were devoid of foreign body reaction but exhibited fibrosis, and GS showed vascularization. There was no significant difference in aqueous or uveoscleral flow. Preoperative morning IOP was 23.7 ± 2 mm Hg, and evening IOP was 26.5 ± 2 mm Hg (P = 0.000). Morning IOP was decreased through 15 weeks and evening IOP through 8 weeks in all groups. The morning IOP decrease was most profound at 15 weeks in PS (41%) compared to GS (18%). Antifibrotics initially enhanced but eventually diminished shunt performance. At 15 weeks, thickness of scleral fibrosis was greater in GS (246 ± 47 μm) and PS (188 ± 47 μm, P = 0.285) compared with GS+MMC (109 ± 26 μm, P = 0.023 to GS) and PS+MMC (48 ± 30 μm, P = 0.028 to PS).

CONCLUSIONS

In a rabbit model, suprachoroidal polypropylene and gold shunts allow access to a new drainage pathway with different IOP profiles that can be modified with antifibrotics.

Tracking in Anatomic Pathology

Bar code–based tracking solutions, long present in clinical pathology laboratories, have recently made an appearance in anatomic pathology (AP) laboratories. Tracking of AP “assets” (specimens, blocks, slides) can enhance laboratory efficiency, promote patient safety, and improve patient care. Routing of excess clinical material into research laboratories and biorepositories are other avenues that can benefit from tracking of AP assets. Implementing tracking is not as simple as installing software and turning it on. Not all tracking solutions are alike. Careful analysis of laboratory workflow is needed before implementing tracking to assure that this solution will meet the needs of the laboratory. Such analysis will likely uncover practices that may need to be modified before a tracking system can be deployed. Costs that go beyond simply that of purchasing software will be incurred and need to be considered in the budgeting process. Finally, people, not technology, are the key to assuring quality. Tracking will require significant changes in workflow and an overall change in the culture of the laboratory. Preparation, training, buy-in, and accountability of the people involved are crucial to the success of this process. This article reviews the benefits, available technology, underlying principles, and implementation of tracking solutions for the AP and research laboratory.

Accounting for the professional work of pathologists performing autopsies

CONTEXT

With an increasing trend toward fee-code-based methods of measuring the clinical professional productivity of pathologists, those pathologists whose clinical activities include the performance of autopsies have been disadvantaged by the lack of generally accepted workload equivalents for autopsy performance and supervision.

OBJECTIVE

To develop recommended benchmarks to account for this important and often overlooked professional activity.

DESIGN

Based on the professional experience of members of the Autopsy Committee of the College of American Pathologists, a survey of autopsy pathologists, and the limited additional material available in the literature, we developed recommended workload equivalents for the professional work associated with performing an autopsy, which we elected to express as multiples of established Current Procedural Terminology codes.

RESULTS

As represented in Table 3 , we recommend that the professional work associated with a full adult autopsy be equivalent to 5.5 × 88309-26. Additional professional credit of 1.5 × 88309-26 should be added for evaluation of the brain and for a detailed clinical-pathologic discussion. The corresponding value for a fetal/neonatal autopsy is 4.0 × 88309-26.

CONCLUSION

Although we recognize that autopsy practices vary significantly from institution to institution, it is hoped that our proposed guidelines will be a valuable starting point that individual practices can then adapt, taking into account the specifics of their practice environment.

Custom software development for use in a clinical laboratory

In-house software development for use in a clinical laboratory is a controversial issue. Many of the objections raised are based on outdated software development practices, an exaggeration of the risks involved, and an underestimation of the benefits that can be realized. Buy versus build analyses typically do not consider total costs of ownership, and unfortunately decisions are often made by people who are not directly affected by the workflow obstacles or benefits that result from those decisions. We have been developing custom software for clinical use for over a decade, and this article presents our perspective on this practice. A complete analysis of the decision to develop or purchase must ultimately examine how the end result will mesh with the departmental workflow, and custom-developed solutions typically can have the greater positive impact on efficiency and productivity, substantially altering the decision balance sheet. Involving the end-users in preparation of the functional specifications is crucial to the success of the process. A large development team is not needed, and even a single programmer can develop significant solutions. Many of the risks associated with custom development can be mitigated by a well-structured development process, use of open-source tools, and embracing an agile development philosophy. In-house solutions have the significant advantage of being adaptable to changing departmental needs, contributing to efficient and higher quality patient care.

Pathology informatics fellowship retreats: The use of interactive scenarios and case studies as pathology informatics teaching tools

Background:

Last year, our pathology informatics fellowship added informatics-based interactive case studies to its existing educational platform of operational and research rotations, clinical conferences, a common core curriculum with an accompanying didactic course, and national meetings.

Methods:

The structure of the informatics case studies was based on the traditional business school case study format. Three different formats were used, varying in length from short, 15-minute scenarios to more formal multiple hour-long case studies. Case studies were presented over the course of three retreats (Fall 2011, Winter 2012, and Spring 2012) and involved both local and visiting faculty and fellows.

Results:

Both faculty and fellows found the case studies and the retreats educational, valuable, and enjoyable. From this positive feedback, we plan to incorporate the retreats in future academic years as an educational component of our fellowship program.

Conclusions:

Interactive case studies appear to be valuable in teaching several aspects of pathology informatics that are difficult to teach in more traditional venues (rotations and didactic class sessions). Case studies have become an important component of our fellowship's educational platform.

Overcoming the Limitations of Integrated Clinical Digital Imaging Solutions

Context.—Digital imaging is increasingly common in medicine. Vendors of anatomic pathology software have addressed this need by integrating digital image acquisition and storage into their products. While offering a number of advantages, these solutions cannot be easily adapted to accommodate the existing work flow for many pathology departments.

Objective.—To describe a novel solution that maintains the advantages of these integrated solutions but offers many additional flexibilities, making it more compatible with the work flow in most clinical departments.

Design.—This solution involves separating the image acquisition step from the image archiving process and creating dual-image storages for greater usability. Software needed to deploy this modular and scalable solution is described.

Results.—We have deployed this solution at our institution and achieved widespread acceptance and use, with minimal training required. Deployment in the surgical pathology gross specimen room, in particular, has resulted in a significant increase in the number of photographs taken and the percentage of cases documented photographically.

Conclusions.—The complete clinical digital imaging solution described herein is an effective, scalable solution for pathology imaging at a departmental level. Although developed and implemented in an anatomic pathology department, the method described is generally applicable to digital imaging in any large multiuser environment.

Glandular patterns in a thyroid carcinoma with insular and anaplastic features: a case with possible implications for the classification of thyroid carcinomas

We describe the case of a 33-year-old woman with a thyroid carcinoma showing poorly differentiated (insular), anaplastic, and glandular features, the latter with extensive clear cell changes. Grossly, the well-circumscribed tumor nodule measured 3.6 cm in maximum dimension and was confined to the thyroid. Microscopically, the majority of the tumor was composed of well-defined "insular" nests showing microfollicular formation, high mitotic activity, and areas of necrosis. Other regions, as well as the intervening stroma of the insular nests, were characterized by highly atypical and pleomorphic stromal cells, extensive necrosis, and malignant cartilaginous nodules. Approximately 30% of the tumor was composed of diffuse glandular formations, each of which were lined by elongated, simple columnar cells with basally situated, mildly pleomorphic nuclei, clear supranuclear, periodic acid-Schiff + (and diastase sensitive) cytoplasm, empty lumens, and no myoepithelia or basement membranes. Immunohistochemically, the glandular elements displayed diffuse and strong positivity for thyroid transcription factor-1, bcl-2, and CAM 5.2, sparse positivity for thyroglobulin and Ki67, and diffuse but weak positivity for p53. Calcitonin was negative throughout the tumor. Karyotypic analysis of a primary culture showed a complex hypertriploid karyotype including structural abnormalities of chromosomes X, 1, 4, 6, 9, 13, and 14 in the majority of cells examined. This composite of histologic findings, especially the glandular patterns, is unusual and their prognostic significance is unclear. The patient is alive with no evidence of tumor recurrence or metastasis at 5 months follow-up. Overall, the morphologic and immunohistochemical properties of the glandular component suggests that they are less differentiated than well-differentiated carcinomas and are probably more differentiated than the insular component. This case supports the theory that the various primary carcinomas of the thyroid may represent points along a spectrum rather than distinct entities.

Clonal analysis of cutaneous fibrous histiocytoma (dermatofibroma)

BACKGROUND

Dermatofibroma (DF) or cutaneous fibrous histiocytoma is a common benign fibrohistiocytic lesion involving the dermis and subcutis. Histologically, it is subclassified into fibroblastic and histiocytoid forms. Its histogenesis is controversial. While often referred to as a neoplastic process, definite evidence of neoplasia in DF has been lacking. Alternatively, some authorities have suggested that DF is a fibrosing inflammatory process. Diagnostically, the most important question faced is the distinction from dermatofibrosarcoma protuberans (DFSP). Misdiagnosis can occur, as the early phase of DFSP can simulate DF, particularly the deep and cellular forms of DF.

METHODS

To address this issue, and to investigate whether DF is in fact a neoplasm, we evaluated 31 examples of DF of various histological types in female patients and assessed clonality by analyzing X-chromosome inactivation as indicated by the methylation status of the androgen receptor gene (HUMARA). Representative cases of DFSP were analyzed for comparison.

RESULTS

Among the selected 31 cases of DF, 24 cases provided intact DNA and informative polymorphism at the AR alleles, including one case of recurrent deep fibrous histiocytoma. Among these 24 cases, randomly inactivated AR alleles were observed in 17 cases including a deep, recurrent fibroblastic DF. A non-random inactivation at AR alleles was observed in seven cases, of which six cases showed either typical histiocytoid form of DF (four cases) or mixed cell types with predominant histiocytoid cell type (two cases). One fibroblastic DF also showed a monoclonal pattern. HUMARA analysis of DFSP revealed non-random inactivation of polymorphic AR alleles.

CONCLUSIONS

These findings suggest that DF is a heterogeneous process. Monoclonal genotype was found in DFs with histiocytoid or mixed type with predominant histiocytoid features, suggesting that histiocytoid cells probably represent the neoplastic component. The fibroblastic form of DF may represent a reactive fibroblastic proliferation. Alternatively, it may represent a true neoplasm whose neoplastic cell type has been obscured by prominent reactive fibroblastic component.

Factors affecting autopsy rates, autopsy request rates, and autopsy findings at a large academic medical center

Autopsy rates continue to decline in the United States. To assess the impact of various objective factors (time of death, day of death, age at death, patient gender, clinical service, and length of hospital stay) on the autopsy request rate, autopsy rate, successful request rate, and percentage of cases in which the autopsy examination added to or altered the clinical assessment of the patient, we prospectively studied all hospital deaths at a major academic medical center for the 3-year period from 1996 through 1998. The autopsy rate decreases significantly with patient age, both because of a decreasing request rate and because family members are less likely to grant permission. An autopsy is less likely to be requested for deaths in the emergency department or on general surgery services and most likely to be requested for fetal, medicine, cardiothoracic surgery, and pediatric deaths. Families more commonly grant permission for autopsy on fetal deaths, pediatric deaths, and emergency department deaths. Forty percent of autopsies reveal significant information about the patient's death beyond what was known premortem. This is least frequent among the fetal deaths, but relatively constant for adults of all ages. Patients who die in the emergency department are most likely to have significant unexpected findings at autopsy. Increasing the request rate for adult patients who die in the emergency department and on the medicine services will result in the greatest increase in information learned from autopsy.

Quality improvement on an academic autopsy service

CONTEXT

Autopsy rates continue to decline in the United States.

OBJECTIVE

Although many of the causes of this decline are external to pathology departments, we hypothesized that intradepartmental efforts to improve the quality of the service we provide to our clinical colleagues could increase our autopsy rate.

METHOD

We developed a multifaceted quality improvement program for our autopsy service aimed at increasing the visibility of the service, improving the service's reporting, and increasing the amount and quality of data available from the service.

SETTING

A large academic medical center that performs approximately 250 autopsies each year.

RESULTS

After implementation of our quality improvement program, the decline in our autopsy rate has not only stopped, but rates have even begun to increase. Additionally, physician satisfaction surveys conducted before and after implementation of our quality improvement initiatives showed an across-the-board improvement in clinician perception of the service.

CONCLUSION

Pathologists can and should be proactive in addressing the declining autopsy rate, rather than viewing it as someone else's problem or hoping that someone else will protect this important quality assurance tool for medical care.

Implementation of a practical digital imaging system for routine gross photography in an autopsy environment

BACKGROUND

The autopsy environment places stringent requirements on a digital imaging system. These requirements must be addressed if the system is to be functional, easy to use, and reliable.

DESIGN

After clearly defining the requirements for such a system, we implemented routine digital imaging in a busy academic autopsy suite.

RESULTS

The new technology was immediately accepted by both the resident staff and the technical staff. Although a 35-mm camera was always available for traditional photography, it was rarely used. An interesting side effect of implementing digital imaging was a nearly twofold increase in the number of images taken per autopsy case. The requirements, features, and utility of a digital imaging system are discussed.

CONCLUSION

Digital imaging in an autopsy environment can be both practical and cost-effective. It provides many advantages over traditional 35-mm photography and can be the first step toward numerous additional improved services.

Immunohistochemical distinction of ocular sebaceous carcinoma from basal cell and squamous cell carcinoma

BACKGROUND

Diagnosis of sebaceous carcinoma of the periorbital region is often delayed. Clinically, this lesion can mimic several inflammatory disorders. Histopathologically, it can mimic either squamous cell or basal cell carcinoma.

OBJECTIVE

To identify an immunohistochemical approach to assist in the diagnosis of periorbital sebaceous carcinoma.

METHOD

The immunohistochemical profiles of several cases of periorbital sebaceous, basal cell, and squamous cell carcinoma were examined.

RESULTS

Although at least focal epithelial membrane antigen (EMA) staining can effectively distinguish sebaceous carcinoma (10 of 11 were positive) from basal cell carcinoma (1 of 16 were positive), most squamous cell carcinomas examined were also focally EMA positive (11 of 14). However, Cam 5.2 reactivity was seen in most sebaceous carcinomas (8 of 11) but no squamous cell carcinomas (0 of 14). In addition, at least focal BRST-1 reactivity was also seen in most sebaceous carcinomas (7 of 11) but no basal cell carcinomas (0 of 16).

CONCLUSIONS

Periorbital sebaceous, basal cell, and squamous cell carcinomas have different immunohistochemical staining profiles; a panel of commonly available antibodies, including anti-EMA, BRST-1, and Cam 5.2, may help distinguish these diseases from each other when that distinction cannot be clearly made by light microscopy alone.

Exclusion of the stomatin, alpha-adducin and beta-adducin loci in a large kindred with dehydrated hereditary stomatocytosis

Defects in stomatin, alpha-adducin, and beta-adducin have been implicated in erythrocyte disorders of cation permeability. We performed linkage analysis of the genetic loci for these proteins in a large kindred with xerocytosis (dehydrated hereditary stomatocytosis). Using polymerase chain reaction-based genotyping techniques, all three loci are excluded as disease gene candidates.

Pigmented villonodular synovitis of the temporomandibular joint: diagnostic imaging and endovascular therapeutic embolization of a rare head and neck tumor

We report a case of pigmented villonodular synovitis involving the temporomandibular joint that presented as a rapidly growing tumor with extension through the skull base into the middle cranial fossa. The case is of interest not only because of the unusual extensive infiltration of this tumor but also because of the role modern diagnostic imaging and endovascular therapeutic techniques played in its diagnosis and management.

Utilization of an 86 bp exon generates a novel adducin isoform (beta 4) lacking the MARCKS homology domain

A novel isoform of beta-adducin has been amplified and characterized from a human bone marrow cDNA library (GenBank #U43959). This isoform arises from the insertion of an 86 bp alternatively spliced and previously unrecognized exon (now termed exon 15) within codon 581 of the human red blood cell beta-adducin sequence. This results in an insertion of 28 novel amino acids. The remainder of the red cell beta-adducin mRNA is then translated in a different reading frame, adding an additional 35 novel amino acids prior to the stop codon. This new isoform, thus, replaces beta 1-adducin sequence after residue 580 with a total of 63 new amino acids. Sequences from genomic clones of the human beta-adducin gene show that this alternate exon is flanked by splice consensus sequences and is appropriately located in the genomic map between exons encoding up-stream and down-stream sequences, thus defining a new exon. The COOH-terminus of this new isoform, which we designate beta 4, lacks a 22 amino acid lysine-rich sequence common to both the human red cell alpha- and beta-adducin subunits and homologous to a highly conserved region in MARCKS, a filamentous actin-cross linking protein regulated by protein kinase C and calcium/calmodulin. beta 4-adducin preserves a previously identified calmodulin binding domain. PCR analysis indicates that this new beta-adducin isoform is expressed in fetal brain and liver, bone marrow, and NT-2 (neuroepithelial) cells, but is not detected in several other tissues. We anticipate that this new beta 4 isoform of beta-adducin will display unique and tissue-specific functional properties.

Neuronal loss from the subthalamic nuclei in a patient with progressive chorea

We present a case of an 80-year-old man who developed a seizure disorder at age 66 and was treated with chronic phenytoin. In the last 3 years of his life, he developed multiple neurological deficits, including bilateral chorea, ataxic gait, sensory neuropathy, and progressive dementia. After death from pneumonia, autopsy examination of the patient's brain was most remarkable for a selective loss of neurons from both subthalamic nuclei and Purkinje cell loss in the cerebellum. This pattern of injury is consistent with a toxic process and does not fit previously characterized pathological syndromes known to be associated with movement disorders or dementia or both. Phenytoin has been shown to cause choreiform movements, peripheral neuropathy, and cognitive decline in some patients, but the pathological basis for these changes has not been elucidated. The patient's chorea was very likely the result of neuronal loss in the subthalamic nuclei, but causes for his dementia and neuropathy were not found. The pathological findings may represent either an unusual form of chronic phenytoin toxicity or a previously undescribed primary degenerative brain syndrome.

Reduced alpha-catenin and E-cadherin expression in breast cancer

BACKGROUND

The expression of the homotypic cell adhesion protein, E-cadherin, is reduced in many types of cancer. The loss of this protein may be associated with metastasis because alteration of its function is required for invasion in vitro, and decreased expression has been associated with more aggressive tumor behavior in vivo. It is likely that the loss of downstream effector elements in the cadherin adhesion cascade may also disrupt cell-cell interactions and thereby promote invasion, but direct evidence for this has been lacking. One such effector element is alpha-catenin, a cytoplasmic protein related to vinculin that is associated in vivo with E-cadherin.

EXPERIMENTAL DESIGN

In the present study, antibodies prepared to recombinant human alpha-catenin and recombinant human E-cadherin have been used to explore by immunocytochemistry the steady state levels of these proteins in a series of 26 cancers of the breast.

RESULTS

The expression of alpha-catenin was reduced or lost more frequently (81% of cases) than was the expression of E-cadherin (63% of cases). Cases with absent E-cadherin expression uniformly lacked alpha-catenin. Eight of the 26 patients (31%) had known metastatic disease at the time of biopsy; yet, all patients with normal alpha-catenin staining in their tumors were free of known metastatic disease (four patients).

CONCLUSIONS

Together with previous data on E-cadherin, these results suggest that reduced steady state levels of alpha-catenin may be a sensitive marker for disturbances in the adhesive function of the junctional complex and suggest that failure of at least one component of the cadherin-mediated cell-cell adhesion cascade is a common feature of breast, and presumably other, epithelial tumors.

Cardiogenic shock due to progression of cutaneous T-cell lymphoma

We describe a patient with progressive cutaneous T-cell lymphoma (CTCL) and development of subacute cardiac failure. Symptomatic lymphomatous involvement of the heart may be more common in patients with CTCL than in other lymphomas because the former is more likely to be associated with circulating tumor cells and hematogenous spread to the myocardium. No single symptom or sign is highly predictive of cardiac involvement, but unexplained tachyarrhythmias, conduction disturbances, low voltage on ECG, and unexplained cardiac enlargement should arouse clinical suspicion. Although echocardiography may be helpful in suggesting cardiac involvement, endomyocardial biopsy should be considered in patients with a reasonable chance of responding to chemotherapy or radiation.

Identification of functional regions on the tail of Acanthamoeba myosin-II using recombinant fusion proteins. II. Assembly properties of tails with NH2- and COOH-terminal deletions

We used purified fusion proteins containing parts of the Acanthamoeba myosin-II tail to localize those regions of the tail responsible for each of the three steps in the successive dimerization mechanism (Sinard, J. H., W. F. Stafford, and T. D. Pollard. 1989. J. Cell Biol. 107:1537-1547) for Acanthamoeba myosin-II minifiliment assembly. Fusion proteins containing the terminal approximately 90% of the myosin-II tail assemble normally, but deletions within the last 100 amino acids of the tail sequence alter or prevent assembly. The first step in minifilament assembly, formation of antiparallel dimers, requires the COOH-terminal approximately 30 amino acids that are thought to form a nonhelical domain at the end of the coiled-coil. The second step, formation of antiparallel tetramers, requires the last approximately 40 residues in the coiled-coil. The final step, the association of two antiparallel tetramers to form the completed octameric minifilament, requires residues approximately 40-70 from the end of the coiled-coil. A region of the tail near the junction with the heads is important for tight packing of the tails in the minifilaments. Divalent cations induce the lateral aggregation of minifilaments formed from native myosin-II or fusion proteins containing a nonmyosin "head," but under the same conditions fusion proteins composed essentially only of myosin tail sequences with very little nonmyosin sequences form paracrystals. The region of the tail necessary for this paracrystal formation lies NH2-terminal to amino acid residue 1,468 in the native myosin-II sequence.

Acanthamoeba myosin-II minifilaments assemble on a millisecond time scale with rate constants greater than those expected for a diffusion limited reaction

We have shown previously that Acanthamoeba myosin-II minifilaments assemble by three successive dimerization steps, forming, progressively, monomers, antiparallel dimers, antiparallel tetramers, and finally the full size octameric minifilament (Sinard, J. H., Stafford, W. F., and Pollard, T. D. (1989) J. Cell Biol. 109, 1537-1548). In the current study, we investigate the kinetics of the assembly of these minifilaments, initiating assembly by the rapid dilution of salt in a stopped-flow light scattering apparatus. The majority of the reaction is completed within 50 ms and is greater than 90% completed within 1 s. Assembly data over a greater than 6-fold myosin concentration range can be fit using the successive dimerization mechanism with a single set of rate constants. Second order rate constants for the initial steps in the assembly reaction exceed 10(8) M-1 s-1, and equilibrium dissociation constants predict a very low critical concentration, consistent with previous data. Other possible assembly mechanisms do not adequately fit all of the available data. Filament disassembly at 300 mM KCl is even more rapid, and there is both an increase in the dissociation rate constants and a decrease in the association rate constants with increasing KCl. Aggregation of minifilaments induced by Mg2+ is much slower and takes many minutes to reach equilibrium.