Early Validation Studies
Four Validation Studies
Dean A. Cole, PhD
Los Alamos National Laboratory
David C. Moody, PhD
Los Alamos National Laboratory
L. Edward Ellinwood, MD, PhD
St. Mary’s Hospital
M. Gerard Klein, MD
St. Mary’s Hospital
Completed 1988 – 1992
St. Mary’s Hospital and the Isotope and Nuclear Chemistry Division of the Medical Radioisotopes Research Program at Los Alamos National Laboratory conducted four preliminary studies prior to application of a patent for 5,10,15,20-tetrakis (4-carboxyphenyl) porphine (TCPP). Two of these studies were designed to determine whether TCPP would localize in neoplastic lung cells from uranium miners. The first study was designed to determine whether TCPP would localize in neoplastic sputum cells from uranium miners and experimental conditions which result in the best localization of TCPP in neoplastic sputum cells. The second study examined the localization of TCPP in different types of lung cancer (squamous cell, small cell and adenocarcinoma) and the ability of TCPP to diagnose lung cancer in different patients in a blind study with different types of lung cancers.
In the initial experiment, the localization of TCPP in sputum samples from two volunteer uranium miners was examined. The first miner, who had confirmed squamous cell lung carcinoma, was designated as the test patient. The second miner was diagnosed as having chronic obstructive pulmonary disease (COPD) with no detectable lung carcinoma. This patient was considered the control.
Five major parameters were investigated in the study:
- The effect of various sputum processing procedures on the uptake of porphyrin in sputum cells;
- The comparison of TCPP to three other porphyrins that are known for their tumor
- uptake;
- The time required for optimal uptake of the porphyrin in sputum cells;
- The uptake of porphyrin in test and control sputum samples; and,
- The verification that TCPP was localizing in malignant sputum cells.
The uptake of porphyrin in sputum cells was evaluated for the number of cells in the sputum samples which fluoresced and the intensity of the porphyrin fluorescence.
The study resulted in the following findings:
- Sputum samples processed with alcohol and carbowax or alcohol (no carbowax) had a larger number of cells free of the mucous than samples processed with phosphate buffered saline (PBS) or unprocessed (raw) samples.
- Sputum samples which had been incubated with TCPP had the greatest number of cells fluorescing (greatest porphyrin uptake) as compared with the three other porphyrins studied. The fluorescent intensity (brightness) of the cells that had localized porphyrin was also greater in samples processed with TCPP than in samples processed with the other porphyrins.
- When sputum cells were incubated with porphyrin for 24 hours and then washed with PBS, the porphyrin remained attached to the cells.
- When porphyrin uptake in the test and control samples was compared, the porphyrin uptake in the control cells was considerably lower than the uptake in the test samples. In addition, the fluorescence intensity of the few control cells that did fluoresce was lower than the fluorescence intensity of the test cells. The most dramatic difference between uptake of porphyrin by test and control sputum cells was seen in samples incubated with TCPP.
- After first staining neoplastic cells in the samples using the PAP, researchers re-examined samples and determined that TCPP fluorescence was seen in every neoplastic cell identified on sputum samples.
The second study evaluated the localization of TCPP in different lung cancer patients with different types of lung cancer. In a blind study, sputum samples from 12 uranium miners with various medical histories were examined for TCPP localization. Eight of the 12 patients had lung cancer, based on medical history and previous PAP staining, and confirmed by biopsy, including three squamous cell lung cancers, three oat cell (small cell) lung cancers, one adenocarcinoma, and one advanced metastatic lymphoma that had spread to the lung. The remaining four patients had been judged free of disease by PAP staining, but had not been biopsied.
The samples were blinded, and researchers who labeled the sputum samples with the TCPP compound were unaware of the patient diagnosis. After labeling with the TCPP compound, researchers counted the number of fluorescent cells in each of three areas of a microscope slide and then multiplied that number by a subjective number between one and four that represented the brightness of the cells that were fluorescent compared with the cells with only “background” levels of fluorescence. Each sample was assigned a fluorescence index (FI). After all samples were scored, the scores were compared to medical histories and data provided by St. Mary’s Hospital.
In 11 out of 12 cases, the samples that had the highest FI were from cancer patients, and the lowest FI were found in controls or normal samples. The one exception was found in a sample that had a FI similar to the known cancer patients, but had been labeled as a control by St. Mary’s. Further PAP stain analysis, prompted by the TCPP results, determined that this subject had been previously misdiagnosed by PAP stain cytopathology, and the patient actually had lung cancer. With this correction, and the addition of one more control or normal sample and one additional cancer sample, the study determined that the FI of sputum from cancer patients was three to eight times greater than from control or normal samples.
Fluorescent Index (FI) of Sputum Samples from Uranium Miners
| Cancer Patients
(Type of Cancer) |
Number of
Patients |
FI (mean) |
| Squamous Cell | 5 | 83.5 |
| Adenocarcinoma | 1 | 68.0 |
| Small Cell Carcinoma | 3 | 42.7 |
| Metastatic Lymphoma | 1 | >150.0 |
| Non-Cancerous | 4 | 15.2 |
In this blinded study, the TCPP compound revealed 100% specificity (4/4) and 100% sensitivity (10/10). In contrast to the TCPP results, the PAP stain had 75% sensitivity, in that three out of four PAP stain negatives were confirmed as patients without lung cancer.
Statistically, the odds of receiving a correct diagnosis for all patients by pure chance are 1 in 4,000. When the additional two samples are included in the statistical analysis, the odds against getting all 14 samples correctly diagnosed is 16,000 to 1.
A third and fourth study were completed to determine whether TCPP would also localize in human lung cancer cells grown in tissue culture and whether flow cytometry could be used to identify lung cancer cells which have localized TCPP (TCPP fluoresces at 650 nm). The third study examined the localization of TCPP in a human squamous lung cancer cell line and determined that squamous carcinoma cells grown in culture localize TCPP and that these cells can easily be detected with flow cytometry. Researchers also determined that the auto fluorescence of control cells (cells not exposed to TCPP) was negligible when compared to the florescence of cells exposed to TCPP.
The fourth study examined a human small cell (oat cell) lung carcinoma cell line and a normal human lung epithelial cell line for TCPP uptake with flow cytometry. Factors that may influence TCPP uptake in these cells, such as different fixative procedures and TCPP incubation times, were also examined. Results of this study demonstrated that TCPP was localized in the small cell carcinoma cell line; however, the TCPP concentration in this cell line was approximately two to three times lower than the TCPP concentration in the squamous cell line. The TCPP concentration in the normal lung cells was slightly above background fluorescence (auto fluorescence), or many times lower than the TCPP concentration in the neoplastic cell lines examined. The study also determined that a 24-hour incubation period for cells in TCPP was more than sufficient and fixing cells prior to TCPP exposure resulted in greater localization of TCPP in lung cancer cells. U.S. Patent # 5,162,231 (Method of using 5,10,15,20-tetrakis (carboxyphenyl) porphine for detecting cancers of the lung) was awarded based on these studies.
Summary Results of Four Studies
TCPP sensitivity was 100% (46/46) in all studies combined; it is 95% certain that the sensitivity is at least 91.2% using a one-sided exact Blyth-Stoll-Casella calculation. TCPP specificity was 100% (21/21); the 95% lower bound is 86.7%.
It is concluded that the sensitivity and specificity are sufficiently high to proceed with a feasibility study to assess independent subgroups of normals, smokers, COPD, and lung cancer patients.
TCPP sensitivity and specificity of the test appears better than the PAP test, which was the “Gold Standard” of cancer diagnostic tests at the time of research testing, which has less than ~65% sensitivity and ~65% specificity with lung cancer.